I've been working on this for a while, it's harder than it should be but it's certainly possible.
The place that helped the most was MSFN Unattended, the documentation there takes you through this kind of stuff step-by-step. They also have forums and driver packs that will help you install to almost any system you can imagine without needing extra driver disks, the BTS DriverPack is the greatest driver collection I've ever seen.
To make your life easier, nLite will automate a bunch of the steps you'll want to use, plus it can strip out a bunch of stuff that you don't want in your install anyway.
When we do so, we'll disclose this to users, just as we already do in those rare instances where we alter results in order to comply with local laws in France, Germany and the U.S.
I don't remember ever seeing that Google filters its results in the U.S. I checked on their TOS and didn't see anything (other than Google disclaims any and all responsibility or liability for the... legality... of information or material displayed in the GOOGLE SERVICES results.), is this talking about C&D orders from copyright holders and the like?
Great that you compare it to "rebooting a computer" because we really have very little idea how ECT works. The brain doesn't do anything akin to rebooting, so this analogy is completely worthless.
I don't have a better reference, but the aforementioned Wikipedia article states "Following the seizure, there is a short period of time during which cortical electrical activity in the brain ceases and an EEG reading is flat," which sounds an awful lot like a reboot to me.
It does permanent changes to the brain and we really have little clue as to what kind of changes. If you want an analogy, maybe "scrambling the hard drive" would be a better one, but even then we really have no clue as to how it works.
It does change the brain, but not like "scrambling the hard drive" as you call it. Again, based on my understanding, I'd say a closer analogy may be reinstalling your OS, but it could be as simple as rolling back whatever changes have been put in place since your last known-good system, and then rebooting. It could even be seen as the ultimate wetware buffer-overflow, it overwrites your brain "code" with crap and you have to shut down and restart.
The fact that it's no longer violent because people are anesthetized when it's done is really beside the point.
Actually, that's entirely my point, but to each their own.
how does shocking something and causing intense and long lasting pain and anguish make them better?
It's not supposed to cause "intense" or "long lasting pain", it's supposed to be administered under anaesthesia so you don't even know what happened. The concept of ECT is sort of like a reboot.
Your computer (like your body) may run fine for a while, it may even go to sleep and wake back up and go on running normally. Eventually you may come across some quirky behaviour (mental disorder) that you can't fix with patches (surgery?) or subsystem resets (drugs?). When all else fails, you reboot your computer (ECT) and everything goes back to normal.
ECT induces a seizure, and your brain sort of shuts down and resets itself. The mechanisms aren't entirely understood, but it works well to treat severe depression.
Electro-Shock Therapy has been portrayed as horrible torture (which it was used for) and has been tried for the treatment of many mental conditions (like schizophrenia and personality disorders) where it does little to no good. It definitely has a shady past, but the modern reality is much more benign and therapeutic.
An OEM heat pipe is nifty, but anyone can buy one to add on.
I'm still waiting for one of Intel/AMD to come out with a chip package that comes with an integrated heatsink or at least heat spreader. Current packaging squishes everything into a 1cm square, then pastes it onto a 3cm PCB that's only that big to get enough I/O pins.
I want to see a chip that integrates a heat spreader directly into the package, so you have some more space to interface for a bigger, more bad-ass heatsink or even heat pipe.
Yes, I do realize that the smaller the chip the larger the yields as far as silicon or SOI goes, but that's actual chip versus the plastic/ceramic/whatever that they use to keep it from being destroyed when you breathe on it. A bigger chunk of that stuff would help a lot, since it could be a single piece instead of one material against another against another, each of which has its own thermal efficiency loss.
Disabling your CD will help, but just keeping it from spinning up is probably fine.
You could also replace your HDD with a flash drive, or better yet, get yourself a hybrid drive. In the past couple days I saw something about IBM trying this, but I don't remember where and I can't find the reference. Just Google for it, there are tons of references out there.
That's a bit of a stretch. Most of us won't find A. Baumannii on our keyboards, it's commonly isolated from the hospital environment. Hospitals are full of nasty stuff that isn't common in the outside world.
I'm not sure, but I think most common are S. Aureus (aka staph) and S. Pneumoniae (aka pneumonia). It's not that you won't find these outside, but the concentration is much higher or the pathogen is much nastier in intensive or acute care settings. In your house you're more likely to find Staph or E. Coli, but they're more benign than their hospital equivalent would be.
As far as badness, pathophys of your baby is roughly the same as other gram-negatives, and it's drug-resistant, like MRSA or some forms of TB. The big guns (new generation fluoroquinolones and similar antibiotics) still work, but it's getting to be a problem.
If you even take the most basic/well studied Lempel-Ziv and Huffman algorithms you'll quickly find cases where each would be preferred over another.
That's sort of the point of this test though, to see which of the general-purpose compressors (GPC) is going to give you the best overall results. Yes, you should use FLAC for WAVs, and probably StuffIt for JPEGs, but what is your best choice if you're going to have just one, or just a few? I don't want 200 different compressors for 200 different content types, I want one.
As a matter of practicality, right now you need zip or gzip, and bzip2 is gaining ground. If you're going to create new content, you should offer both bz2 and zip. In the future, maybe you should use 7z or sit instead, it depends on the rate of adoption. Personally, I don't think zip will ever die.
And since different algm's identify different patterns in the file their compressing, certain files will be compressed better by different algorithms and do much worse on the next file. Besides, we're not even getting into any discussion of lossy/lossless algm's here. (Think jpeg vs bmp).
Generally, you will pick a special-purpose compressor for lossy compression, and a GPC for lossless compression. Your audio compressor will probably be MP3 or OGG, your images will probably be JPG, videos will be MPG. It's not efficient to use MP3 compression on your images, it's designed with different constraints. Either for the same bitrate the image is much worse quality, or for the same quality the file will be much larger than necessary. The same goes for lossless compressors too, FLAC works much better than ZIP on audio data, but I would bet if you used a BMP file as the source for compression FLAC would probably be bad and ZIP would probably be average.
If you want to compress 300 files of various types, you need a GPC. That doesn't mean that the GPC doesn't have special-purpose algorithms built into it, it just means that on-average it will perform better than a special-purpose compressor.
Kolmogorov complexity, or at least an estimate thereof, is what you're talking about. For any specific dataset, the Kolmogorov complexity is the minimum size of compressed data + decompressor. It can't be calculated, but it is a measure of performance for any combination of compressor and dataset. For WAVs, you will probably see this: K(FLAC, WAVs) < K(GPC, WAVs)
However, for an evenly-distributed general dataset of generic binary files, TXT, JPG, PDF, TIF, PNG, MP3, WAV, and MPG, you will probably find that for any SPC (special-purpose compressor for any of the individual data types): K(GPC, dataset) < K(SPC, dataset)
The pyramids are still standing. Stonehenge is too. This clock is made out of stainless steel and monel, a "nickel-copper alloy" that is known to be corrosion resistant. The final version is expected to be made of the same, plus some bronze and other long-wearing substances. The overall design principles of the Long Now clocks will make them physically durable, it seems like mechanical longevity is going to be the least of their problems.
The anthropologic aspect of this project is going to be the most difficult, simply because society is a factor. The rise and fall of civilizations happens much more often than the rise and fall of material objects. We can still recover bronze-age artifacts (circa 5000 years old), and even some from the stone age (anywhere from 8,000 to 30,000 years old), but we have very little information on what the societies were like. Most of what we have is just a guess.
The good news is that those same design principles that make it physically longstanding address these problems from a sociologic / anthropologic POV also.
Maintainability - The clock should be maintainable with bronze-age technology
1. Already done, stock demand sheets are generated daily as FYIs and warehouse is reconfigured as necessary. Don't have to worry about finding slots, stocking and picking lists are generated based on these sheets and on inventory levels. I'm quite proud of the Ops folks on some of this, they use some really clever, simple technology to improve their efficiency by leaps and bounds.
2. Not sure if this makes a difference, most locations are load overnight, drop overday, repeat ad infinitum. A faster or slower truck would still be cycle-limited.
3. If movement space is limited then this could be a much bigger benefit, but I think this is done inherently since we can't put 550 cubic feet of stock in a 500 cubic foot trailer, we automatically route it to a bigger truck or split to two trucks, whichever makes more sense.
4. Done, to some extent. Product Buyers account for seasonal swings in demand (Cinco De Mayo = lots of salsa, Thanksgiving = lots of turkey), but I don't know how specific they are in their forecasting.
5. This might be the best option for my wandering mind, there are lots of reports that have to be run overnight because they take too long, running on a desktop cluster could make relatively real-time reporting a reality. Really!
This would be a good one to test, too, since each group would be willing to give up idle time to process their own reports that much quicker. I think we even have a cluster-enabled database license for some other one of our apps, so we have the personnel in-house to support it.
We actually have people that do each of these things.
1. Our Buyers have to track stock versus demand to keep products on the shelf but only at the minimum level possible, otherwise we're storing more than we need and that extra storage costs us extra money.
2. Buyers have to do this also, in relation to #1 and supply-versus-demand across multiple markets.
3. Routing people do this, but they do it manually AFAIK. For this, as for some others on this list, reality becomes a problem.
Logistically, it makes the most sense to re-optimize your drop schedule every night to account for new customers, larger demand, changes in traffic flow, etc.
Reality says that Driver A knows these twelve stops and has run the same route for two years, so he/she knows the intricacies of those customers and/or that route.
4 and 5. AFAIK, these are dealt with on a one-off basis, when they happen someone fixes them. I'm sure there's some global strategy on how these are tracked, but I haven't investigated it.
6. This is handled by our transportation and operations people. I'd bet hard cash that it isn't tracked as well as it could be, but there have been a lot of incremental process improvements globally since I started working with this company 7 years ago, so we're probably in the top 10-15% of our business in efficiency.
7. Again, handled by buyers, and again part of the supply-demand problem they face in #1 and #2.
Overall, every metric referenced on this list can come out of our enterprise data warehouse, so we CAN track all of this. I even think that we DO track most of it, and adjust our business processes accordingly.
The only downside to this, from my wishful perspective, is that most of these are one-shot deals. I can write the query to give me any of these current metrics along with past performance, then I can process the result set with just my PC in 5 minutes. There's no magic to it.
(From a user's viewpoint though, this is practically frickin' nirvana!)
Thus the reason that any distributed project must contribute directly to the business. If that weren't the case, SETI would be running on everyone's desktop already because it's a pretty screensaver and it's completely worthless from a business perspective.
It would be great to do charity computing, but that idea has already been nix'd. I had that in my original posting, but it got edited out before the final submission:
I asked about charitable computing and that was shot down. "They" did say that a project that would contribute directly to the business (read: contribute MONEY) would be allowed, as long as some minor points were addressed.
Well, they must need the processing power for something. After all they didn't buy 20 servers and 400 workstations just to look cool. Maybe they need the processing power for suff like groupware, inventory control, accounting, human resources and general "work stuff".
For a desktop box though, how much processing time do you use in a day? I don't know the equivalent for Linux (maybe uptime?), but check task manager if you run Windows.
For my desktop PC, even when I was doing a LOT of Access database work (as opposed to a dedicated db where it has its own server), I was running at about 7 hours of idle processing time in a 9-hour day. There are maybe 10 people in the company that actually use their computers to the fullest, everyone else runs e-mail, Excel and a mainframe client.
Servers are mostly in the same boat, although we don't have nearly as much overkill for them as we do for desktops. The specs for most of our apps require a dedicated server with the appropriate redundant hardware, so even if there are only four people that run on our contract management server, we must have separate, dedicated hardware for them. As of Q4 last year we started running VMWare virtual servers to consolidate that somewhat, but for the time being we have more horsepower than we could use in a decade.
By the way, it seems that you don't have enough to do at your new job. Perhaps you should ask your boss what would be the best way to allocate company computing resources and your leet admin skillz. I'm sure he wouldn't have thought of Slashdot.
I have enough to do that I can't comment on/. at 4:16PM EDT, I have to wait until I'm done at 7:30.
That's bizarre! I've never heard of this guy, nor seen his site, but there's the death squad in all its glory! I wonder if this has been featured on Halfbakery too?
I always wanted to round up a kindergarten class worth of children to act as my solar death squad. Here's how it would work:
1. Give each kid their own shiny little mirror with a post-it note stuck on it to block the shiny part 2. One at a time, have them remove the post-it, aim their mirror to reflect the sun upon some point, then re-post-it. 3. Once everyone is aimed (30 kids or so), have them all remove their post-its at once, instantly creating a plasma-hot ball of fire at the point of focus, incinerating your enemies with the might of a kindergarten class.
Has anyone else had this idea too, or am I the only weirdo around here?
I'm looking into getting maybe three or four of these mini-ITX systems to do stereo vision processing for a DARPA run. They're easier to work with than usual embedded controlers (since they're 99.9% x86 PC), and they're much cheaper than even the cheapest laptops out there.
These in particular might not be cheap because they're so new, but a Nehemiah 1Ghz is only £95GBP or $171USD (guess they haven't updated their exchange rates), $760.02 for 4 of 'em including shipping anywhere in the lower 48.
a 42-U rack with 168 processors would draw about 2.5 kilowatts, or about as much power as two hair dryers." This also looks like the basis for a nice car computer.
I know this is News for Nerds and all, but isn't that a bit excessive? I don't think my car needs 168 1GHz processors. (or is that 336 processors?) What's it going to do with that much power?
Slashdot Mirror
"I think there is a world market for maybe five computers." - Thomas Watson, 1943
"640k ought to be enough for anybody" - Bill Gates, 1981
"Consumers don't even need 10 gigabit, why would they want 100 gigabit?" - Guspaz, 2006
I've been working on this for a while, it's harder than it should be but it's certainly possible.
The place that helped the most was MSFN Unattended, the documentation there takes you through this kind of stuff step-by-step. They also have forums and driver packs that will help you install to almost any system you can imagine without needing extra driver disks, the BTS DriverPack is the greatest driver collection I've ever seen.
To make your life easier, nLite will automate a bunch of the steps you'll want to use, plus it can strip out a bunch of stuff that you don't want in your install anyway.
It does change the brain, but not like "scrambling the hard drive" as you call it. Again, based on my understanding, I'd say a closer analogy may be reinstalling your OS, but it could be as simple as rolling back whatever changes have been put in place since your last known-good system, and then rebooting. It could even be seen as the ultimate wetware buffer-overflow, it overwrites your brain "code" with crap and you have to shut down and restart.
Actually, that's entirely my point, but to each their own.
Your computer (like your body) may run fine for a while, it may even go to sleep and wake back up and go on running normally. Eventually you may come across some quirky behaviour (mental disorder) that you can't fix with patches (surgery?) or subsystem resets (drugs?). When all else fails, you reboot your computer (ECT) and everything goes back to normal.
ECT induces a seizure, and your brain sort of shuts down and resets itself. The mechanisms aren't entirely understood, but it works well to treat severe depression.
Electro-Shock Therapy has been portrayed as horrible torture (which it was used for) and has been tried for the treatment of many mental conditions (like schizophrenia and personality disorders) where it does little to no good. It definitely has a shady past, but the modern reality is much more benign and therapeutic.
An OEM heat pipe is nifty, but anyone can buy one to add on.
I'm still waiting for one of Intel/AMD to come out with a chip package that comes with an integrated heatsink or at least heat spreader. Current packaging squishes everything into a 1cm square, then pastes it onto a 3cm PCB that's only that big to get enough I/O pins.
I want to see a chip that integrates a heat spreader directly into the package, so you have some more space to interface for a bigger, more bad-ass heatsink or even heat pipe.
Yes, I do realize that the smaller the chip the larger the yields as far as silicon or SOI goes, but that's actual chip versus the plastic/ceramic/whatever that they use to keep it from being destroyed when you breathe on it. A bigger chunk of that stuff would help a lot, since it could be a single piece instead of one material against another against another, each of which has its own thermal efficiency loss.
Disabling your CD will help, but just keeping it from spinning up is probably fine.
You could also replace your HDD with a flash drive, or better yet, get yourself a hybrid drive. In the past couple days I saw something about IBM trying this, but I don't remember where and I can't find the reference. Just Google for it, there are tons of references out there.
That's a bit of a stretch. Most of us won't find A. Baumannii on our keyboards, it's commonly isolated from the hospital environment. Hospitals are full of nasty stuff that isn't common in the outside world.
I'm not sure, but I think most common are S. Aureus (aka staph) and S. Pneumoniae (aka pneumonia). It's not that you won't find these outside, but the concentration is much higher or the pathogen is much nastier in intensive or acute care settings. In your house you're more likely to find Staph or E. Coli, but they're more benign than their hospital equivalent would be.
As far as badness, pathophys of your baby is roughly the same as other gram-negatives, and it's drug-resistant, like MRSA or some forms of TB. The big guns (new generation fluoroquinolones and similar antibiotics) still work, but it's getting to be a problem.
Doctors and nurses, please wash your hands!
As a matter of practicality, right now you need zip or gzip, and bzip2 is gaining ground. If you're going to create new content, you should offer both bz2 and zip. In the future, maybe you should use 7z or sit instead, it depends on the rate of adoption. Personally, I don't think zip will ever die.
Generally, you will pick a special-purpose compressor for lossy compression, and a GPC for lossless compression. Your audio compressor will probably be MP3 or OGG, your images will probably be JPG, videos will be MPG. It's not efficient to use MP3 compression on your images, it's designed with different constraints. Either for the same bitrate the image is much worse quality, or for the same quality the file will be much larger than necessary. The same goes for lossless compressors too, FLAC works much better than ZIP on audio data, but I would bet if you used a BMP file as the source for compression FLAC would probably be bad and ZIP would probably be average.
If you want to compress 300 files of various types, you need a GPC. That doesn't mean that the GPC doesn't have special-purpose algorithms built into it, it just means that on-average it will perform better than a special-purpose compressor.
Kolmogorov complexity, or at least an estimate thereof, is what you're talking about. For any specific dataset, the Kolmogorov complexity is the minimum size of compressed data + decompressor. It can't be calculated, but it is a measure of performance for any combination of compressor and dataset. For WAVs, you will probably see this:
K(FLAC, WAVs) < K(GPC, WAVs)
However, for an evenly-distributed general dataset of generic binary files, TXT, JPG, PDF, TIF, PNG, MP3, WAV, and MPG, you will probably find that for any SPC (special-purpose compressor for any of the individual data types):
K(GPC, dataset) < K(SPC, dataset)
The anthropologic aspect of this project is going to be the most difficult, simply because society is a factor. The rise and fall of civilizations happens much more often than the rise and fall of material objects. We can still recover bronze-age artifacts (circa 5000 years old), and even some from the stone age (anywhere from 8,000 to 30,000 years old), but we have very little information on what the societies were like. Most of what we have is just a guess.
The good news is that those same design principles that make it physically longstanding address these problems from a sociologic / anthropologic POV also.
Maintainability - The clock should be maintainable with bronze-age technology
Maintainability and transparency:
(emphasis added)
Create multiple copies of your base data and include error recovery codes.
RAR has error recovery features built in, there are also add-on tools to create parity/error correction blocks from whatever data you throw at it.
Of course if multiple copies from different sources are all damaged, you probably have bigger problems than getting your "important" data back.
1. Already done, stock demand sheets are generated daily as FYIs and warehouse is reconfigured as necessary. Don't have to worry about finding slots, stocking and picking lists are generated based on these sheets and on inventory levels. I'm quite proud of the Ops folks on some of this, they use some really clever, simple technology to improve their efficiency by leaps and bounds.
2. Not sure if this makes a difference, most locations are load overnight, drop overday, repeat ad infinitum. A faster or slower truck would still be cycle-limited.
3. If movement space is limited then this could be a much bigger benefit, but I think this is done inherently since we can't put 550 cubic feet of stock in a 500 cubic foot trailer, we automatically route it to a bigger truck or split to two trucks, whichever makes more sense.
4. Done, to some extent. Product Buyers account for seasonal swings in demand (Cinco De Mayo = lots of salsa, Thanksgiving = lots of turkey), but I don't know how specific they are in their forecasting.
5. This might be the best option for my wandering mind, there are lots of reports that have to be run overnight because they take too long, running on a desktop cluster could make relatively real-time reporting a reality. Really!
This would be a good one to test, too, since each group would be willing to give up idle time to process their own reports that much quicker. I think we even have a cluster-enabled database license for some other one of our apps, so we have the personnel in-house to support it.
We actually have people that do each of these things.
1. Our Buyers have to track stock versus demand to keep products on the shelf but only at the minimum level possible, otherwise we're storing more than we need and that extra storage costs us extra money.
2. Buyers have to do this also, in relation to #1 and supply-versus-demand across multiple markets.
3. Routing people do this, but they do it manually AFAIK. For this, as for some others on this list, reality becomes a problem.
Logistically, it makes the most sense to re-optimize your drop schedule every night to account for new customers, larger demand, changes in traffic flow, etc.
Reality says that Driver A knows these twelve stops and has run the same route for two years, so he/she knows the intricacies of those customers and/or that route.
4 and 5. AFAIK, these are dealt with on a one-off basis, when they happen someone fixes them. I'm sure there's some global strategy on how these are tracked, but I haven't investigated it.
6. This is handled by our transportation and operations people. I'd bet hard cash that it isn't tracked as well as it could be, but there have been a lot of incremental process improvements globally since I started working with this company 7 years ago, so we're probably in the top 10-15% of our business in efficiency.
7. Again, handled by buyers, and again part of the supply-demand problem they face in #1 and #2.
Overall, every metric referenced on this list can come out of our enterprise data warehouse, so we CAN track all of this. I even think that we DO track most of it, and adjust our business processes accordingly.
The only downside to this, from my wishful perspective, is that most of these are one-shot deals. I can write the query to give me any of these current metrics along with past performance, then I can process the result set with just my PC in 5 minutes. There's no magic to it.
(From a user's viewpoint though, this is practically frickin' nirvana!)
Thus the reason that any distributed project must contribute directly to the business. If that weren't the case, SETI would be running on everyone's desktop already because it's a pretty screensaver and it's completely worthless from a business perspective.
It would be great to do charity computing, but that idea has already been nix'd. I had that in my original posting, but it got edited out before the final submission:
I asked about charitable computing and that was shot down. "They" did say that a project that would contribute directly to the business (read: contribute MONEY) would be allowed, as long as some minor points were addressed.
For my desktop PC, even when I was doing a LOT of Access database work (as opposed to a dedicated db where it has its own server), I was running at about 7 hours of idle processing time in a 9-hour day. There are maybe 10 people in the company that actually use their computers to the fullest, everyone else runs e-mail, Excel and a mainframe client.
Servers are mostly in the same boat, although we don't have nearly as much overkill for them as we do for desktops. The specs for most of our apps require a dedicated server with the appropriate redundant hardware, so even if there are only four people that run on our contract management server, we must have separate, dedicated hardware for them. As of Q4 last year we started running VMWare virtual servers to consolidate that somewhat, but for the time being we have more horsepower than we could use in a decade.
I have enough to do that I can't comment on
That's bizarre! I've never heard of this guy, nor seen his site, but there's the death squad in all its glory! I wonder if this has been featured on Halfbakery too?
I like the Duck-plunge Mechanical Fountain, that could be a neat waterpark ride.
I always wanted to round up a kindergarten class worth of children to act as my solar death squad. Here's how it would work:
1. Give each kid their own shiny little mirror with a post-it note stuck on it to block the shiny part
2. One at a time, have them remove the post-it, aim their mirror to reflect the sun upon some point, then re-post-it.
3. Once everyone is aimed (30 kids or so), have them all remove their post-its at once, instantly creating a plasma-hot ball of fire at the point of focus, incinerating your enemies with the might of a kindergarten class.
Has anyone else had this idea too, or am I the only weirdo around here?
Well, IIRC, it will dry your hair. And that of your passenger.
Or you could just stick your head out the window and leave the cluster at home, that's what I usually do.
Funny you should mention that...
I'm looking into getting maybe three or four of these mini-ITX systems to do stereo vision processing for a DARPA run. They're easier to work with than usual embedded controlers (since they're 99.9% x86 PC), and they're much cheaper than even the cheapest laptops out there.
These in particular might not be cheap because they're so new, but a Nehemiah 1Ghz is only £95GBP or $171USD (guess they haven't updated their exchange rates), $760.02 for 4 of 'em including shipping anywhere in the lower 48.
a 42-U rack with 168 processors
nevermind...
a 42-U rack with 168 processors would draw about 2.5 kilowatts, or about as much power as two hair dryers." This also looks like the basis for a nice car computer.
I know this is News for Nerds and all, but isn't that a bit excessive? I don't think my car needs 168 1GHz processors. (or is that 336 processors?) What's it going to do with that much power?
I though it was shorn, as in sheared, like what you do to a sheep... Shorn?
I just can't tell with the accent, Shaun and Shorn are about the same in British English.