Slightly off topic, but very closely related, does anyone have stats on what results in more activity, structured scheduled programmed events or free time?
My gut level guess is free time results in much higher activity levels than structured events. Building a snow fort is exhausting compared to warming a bench at a sport. Etc.
It Might be that enforced structured scheduled wii play (OK governor katy pery song on just dance 3 will commence from precisely 1012 to 1015, after which the next scheduled song, like it or not, will be...) is just boring and less energetic than kids just going crazy doing whatever they want.
Thinking back to my youth, when I wasn't chasing dinosaurs, hunting for woolly mammoth, or going fishing for trilobites, if I built a snow fort however I wanted I went crazy building it until I'd practically collapse, but given an "assembly line job" of repeatedly replicating a standardized boring data analysis driven snow fort I'd probably have all joy crushed out of the experience.
Short version is butchering wood in the basement workshop = big fun = high energy but middle school shop class making a birdhouse = no fun, I Fing hate birds = just do the minimum until next class. Wii might be the same way.
Anonymous and Wikileaks please help! Peoples lives depend on results kept hidden by two gangs of crooks, and only wikileaks and/or anonymous can save peoples lives... The government sure as heck won't help the people, the crooks are actively against us... the only people who can save lives are you two. go go go
1) One bunch of crooks makes money by letting people die in mines for higher profit. Evil personified.
2) One bunch of crooks makes money by charging both authors and readers to distribute research funded by taxpayer dollars. Pointless dying intermediaries.
I would love it if both evil groups of crooks get screwed. Anyone got a link to a torrent on the pirate bay yet? Or a.onion tor link, or a i2p link, or... note I'm not looking for links to rickrolls and goatse, only links to the genuine documents (assuming there's no false flag operation where the mining co. themselves release "massaged" data, not outside the realm of possibility)
Are there no superheros anymore? Is this just too difficult for anon and wikileaks? Come on guys, get it in gear and save some lives.
a quick (bash runonallmachines.sh apt-get install foobar) could print out the info to stdout
This seems to be turning into an endless array of "lets find even harder wheels to buggily reinvent". Its NIH all the way down each step of the way... The proper way to do this wrong thing in this situation, is to "apt-get install dish" which is the distributed shell package. Put yer list of hostnames in ~/.dish/hosts and yer options in ~/.dish/options (-p0 and -T 300 are fairly popular...). Any sane person has already figured out ssh keys and ssh agent and sudo all that before hand. Then its just "dish apt-get install foobar"
Also if you've ever tried this unique form of cluster package management, its a fail if apt-get blocks on some debconf questions, or blocks on waiting on input asking if its ok to install dependencies. Also fails if any previous package installation failed awaiting dpkg-reconfigure -a or whatever. And it fails if something went weird during the most recent (perhaps automatic?) apt-get update. Again, you've not figured out how to manually process the logs or manually process failures.
It is true that there exists a way to do package management which is harder and slower and more tedious, and that is to manually by hand ssh into each machine and run each command by hand. But finding a solution one millimeter above the absolute bottom of the barrel, doesn't mean you've found the optimum solution.
You are correct, in that you've just found a way to avoid the staggering and overwhelming job of adding this line to some init.pp file on ye olde puppetmaster:
package { "foobar": ensure => latest }
Seriously, using puppet really IS that hard. We're not talking about writing a Turing machine in sendmail M4 macros here. Puppet was designed to do Exactly what you're trying to do, and do it very well, very easily, and very quickly. So, just do it... There's absolutely nothing wrong with hacks for the sake of doing a hack, or reinventing something to figure out how its done. But when its time to work, use the right tool for the job. Don't install drywall screws with a hammer.
it became cheaper and easier to locate decent beer so people that brewed just to get good beer no longer needed to brew at home
That's like trying to argue that people only cook at home because there are not enough restaurants. "If they'd just open a Thai restaurant around here, then architects could stop putting kitchens in homes". Don't think so...
I'll be honest, several of my experiments in brewing tasted awful, much like some of my cooking experiments would have been best not eaten. The fun is in the experience of making it myself, my way. Its like solving a big puzzle.
The other part is brewing is "big enough work" that it becomes a social activity. You can make a social activity out of the alternative of driving to the liquor store and waiting in the checkout line, but trust me that brewing is much more fun.
The biggest problem I have with microbrews at the store is they sit there long enough under the lights to get old and skunky. Might have been good fresh, not so good after 6 months getting dusty on a shelf. Yuck. Laughably some people think skunky beer is the "new taste" microbrews are aiming for...
I believe beer is the perfect lens through which to examine innovation,
Here's the most important lesson which I bet is either not covered accidentally or maybe intentionally.
I live in a greater-city which used to be the center of American beer brewing. A century or so ago, German immigrants built dozens of medium sized breweries and exported all over the country. Big big names, still around in marketing even today.
All of those jobs, and I mean all, are gone, inside the city. Every last one. Mergers inside the country and international, centralization, downsizing, blah blah, and now we've gone from dozens of breweries to a handful of microbrews, depending on how you want to count Sprecher (in a nearby city) and this brewpub by the local engineering college. A century ago there were dozens of people in my city with the job title "brewmaster" now there is debate but the number seems to hover right around "one" or "zero" depending how picky you want to be.
Similar thing happened in the automotive business, from hundreds of companies a bit over a century ago to just a handful now. Same deal multiple times with computing.
The lesson is that in a Emerging Technology there might be thousands of management and engineering jobs, but eventually its no longer an Emerging Technology then almost ALL of those jobs go away, permanently. If you're a 1 in a 100, maybe you can be a survivor making a long term career out of emerging tech, or if you enjoy perma-unemployment after a real fun 10 year run that'll work, but otherwise, if you see emerging tech, run like hell away, if you care about your family being able to eat and have a roof over their head. Run!
I've heard it said the difference between a homebrewer and a pro is like the difference between a internal skeleton and an exoskeleton, in that a homebrewer sanitizes by putting stuff into the kitchen (or basement) sink, whereas the pro takes sanitizing solution out of the sink and into the apparatus because its so big.
I sell home beer and wine making supplies and ingredients in my hardware store.
As a guy who brewed beer in the past and probably will again in the future, most of the stuff you sell to brewers, you probably don't know about.
I used to buy replacement plastic transfer hoses, copper tubing and handful of strange compression fitting adapters to make my own wort chiller, tubes and hoses to make my own homemade bubbler/vaporlock thingy, etc etc. I purchased all the gear to make what amounts to a remote faucet system on a hose for cleaning. I had the worlds weirdest rube goldberg device to fill bottles. For wine/mead supposedly the most expensive and traditional primary fermenter is a glass carboy, and supposedly the cheapest is a food grade plastic bag (not insecticide treated garbage bag) inside a non-value engineered old fashioned strong metal trash can. Supposedly prices have exploded upward so much that the cheapest durable and watertight primary "couple gallon" fermenter is a standard tropical fish aquarium, although keeping light out and the top sealed must be a huge PITA.
I never bought "normal homebrewing stuff" from a hardware store like yeast and hops, bottle caps for my crimper, whatever. Thats cool that you sell that stuff as I have 4 hardware stores within 5 miles, but my local "homebrew store" was at least an hours drive. In the internet era its more realistic to order online and wait a day or two, than to invest that kind of windshield time.
Speaking of which, can you even buy non-cordless phones these days?
You have to go to walmart, but they do exist.
The real problem is you need a copper line that is homerun to a CO not to a SLC hut. SLC hut battery backup is... not so good, if there is any at all. Supposedly there existed a SLC-96 system 30 years ago that was CO powered off the T-1 repeater supply, but they've only installed fiber SLCs for decades now and the few I've seen the insides of have metered electrical service.
People who only know a little telecom think every copper line is a home run to the CO. People who worked in telecom know that 20 years ago SLC market penetration was at least 1 in 10 residential lines, and now, I would not be surprised if the majority of copper lines are run to a SLC.
It does depend on your neighborhood. If there's a homeless panhandler on the sidewalk, thats urban and probably 100 years old and you probably have copper homerun, unless its a "factory to condo conversion" and the telco put a SLC in. If its a "1950s baby boomer house" like mine then its 50/50 and in fact mine is on a SLC (which was a nightmare to get DSL back in ye olden DSL era). If its a modern mcmansion I guarantee the LEC installed a SLC, they're not going to home run copper all they way from each house to the CO.
LEDs are used to cycling at 120 hz, or I guess 100 Hz in soccer-hooligan-land. This might be an excellent excuse to upgrade to LEDs.
Does anyone in the biz know what they use to drive the LEDs? I'm assuming a simple bridge rectifier and some manner of constant current switcher, but "real genuine streetlights" might do something more exotic, I donno.
Note I'm not interested in how model railroader or RC car builders bias a LED using a simple resistor, I'm quite well informed on that, thanks, I'm looking for real details how real streetlight ckts work.
Of course, the control system required is far more complicated here. I wonder how much energy is consumed in producing and maintaining the new lampposts, controls, communication network, etc.
I put on my engineers estimating hat and...
I will call 1 KW per lamp. Yes its probably beneath but not much.
Cost per KWh is below 10 cents/KWh for everybody, far below for a major account like my city or the highway dept, but its "In the ballpark" when you add in corruption, management, accounting, overhead in general. So 1 KW lamp costs $0.10 per hour
yearly average is 12 hours per day of lamp and 365 days per year but thats a pain to multiply so we'll call it 10 hours a day and 333.333 or a 1/3 of a thousand days per year, for a third of a thousand of ten tenths bucks per year or simplifying a third of a thousand bucks per year.
I don't see why they'd have to replace the lampposts. The control gear is not going to last 25 years or whatever mechanical timer relay based controls lasted, this is going to be a MS windows installation with no upgrade path other than forklift an entirely new system in, so lets call it optimistically 10 years.
So over the life of the install, you would spend $3333 for a dumb system, so your smart system cannot physically save more than $3K or so, so unless its the usual greenwashing corruption it needs to cost an order of magnitude less, like a couple hundred per pole absolute maximum.
Which, looking at X10 / insteon type stuff is doable, but if you want to use real scada controlled PLCs is probably not doable. It'll be a tough installation.
Also before anyone pipes up, I'm sure the output changes will increase wear and tear on the bulbs, but thats counterbalanced by the cost of electricity increasing over time, so I think it fair to disregard.
One of the problems about dimming lanterns is that the lamp post spacing is all based around the lamps at a certain luminenscence and so dimming may create dark zones, or over bright zones. So some careful analysis will be needed about how the lamps dim and whether they dim uniformly or not.
I can personally verify this zone problem happens with at least some LED lights. I live in a city that has been testing old fashioned LED lights (a couple years, maybe a decade behind China at least) and they're trying to retrofit the LED heads onto existing poles but the heads they chose do not output the same light cones as the old sodium vapor or whatever lamps. So we get bright and dark bands flickering as we drive. I am not epileptic but I know the flashing sets them off and its bad enough to be disturbing to me, so it must be absolutely awful for them. Also the glare from the LED point sources is intensely dazzling to the point of danger whereas the older lamps did not have as high of a peak glare output.
I live in snow country, and the LED streetlamps do get hot enough to melt the snow off them. Must be a hundred watts of luxeon up there. The very well known problem of stop and go lights completely encased in snow and ice does not happen with the LED streetlamps, or at least not the ones we have.
This does not absolutely prove LEDs suck; its entirely possible the laborers who made a living replacing old fashioned light bulbs convinced someone to intentionally install a batch of fail for the test program so they can keep their jobs after we rip out the horrid LEDs.
Intel AppUp Center is an online repository designed for Intel processor-based devices.
Minor correction; its a windows only app store. Does not perform the miracle of running the same executable on mac osx, all linux distros, and windows. Just windows thats all.
libreoffice is available for all those platforms, just not available on the windows only appup
If you only have 30 machines for one event, puppet/cfengine is overboard. Just set up a passwordless SSH key for root (remember NOT to put the private key on the laptops), and just use a simple script to send the same commands to every laptop.
You forgot error recovery and logging. So PC #25 was rebooting while your script ran... does that make the script fail, perhaps silently? Does that mean all the PCs except #25 are OK, or #25-(end of list) fail? How do you know to rerun the script later? How many times do you have to run it by hand to "make sure" it ran on all machines? You can add logging and some sort of retry mechanism... Just remember that those who try not to use puppet, end up rewriting puppet, just takes a long time and painful bugs. apt-get install puppetmaster on the main and apt-get install puppet on the remotes is just too easy to bother writing a clone of puppet out of shell scripts.
Puppet is much like AFS or LDAP or well, practically any service, the first time you set it up you're all "WTF?" and the second time its no big deal. Google for some tutorials, screencasts... Doing a really halfway job writing it yourself is terrifyingly harder than just using puppet.
With all due respect, nothing personal, but the ideas you expressed are completely wrong. Kids need to learn that science is experimenting and debating and arguing and trying things that mostly don't work but sometimes they do. There is no cabal and smart people sometimes disagree, most importantly they disagree in a civilized manner. And getting excited and theorizing and double checking your work and then triple checking your work and lots of sweat and effort and long hours. Initial results are sometimes wrong. Where do errors come from? And sometimes how you deal with "failure" defines who you are, more than how you deal with "success".
Science is not (or should not be) a scholastic endeavor that we should try to make as boring and authoritarian and slow and uninteresting as possible. If anything try to make it the opposite, at least a little bit.
If this whole story makes one kid think, just a little bit, about physics, that makes it OK. This is the best thing thats happened to physics in years.
If science were as flaky as a reality TV show, then I'd support your position because somewhere in between is the greek ideal. But... there's a long way to go before we have to worry about that.
I have to think about this for a second, but I'd assume the neutrinos had a Very accurately measured speed. So maybe the story isn't that the neutrinos go supraluminal, its that fiber optic cable at CERN has a slightly negative velocity factor...
Now you'd think people have measured the length of fiber with a OTDR before (god knows I have enough time) but maybe there is something weird about CERNs fibre, like they had to wrap the slack somewhere and they had a 10 tesla superconductive magnet laying around of a convenient diameter, so... , and no one has ever OTDRd something that stupid before (although I remember doing some stupid OTDR tricks when I was learning in the 90s)
>Fuck it, I'm going to Greece.
>Go to Greece.
>Considering marrying a Greek girl
>Berlusconi is there
You missed a golden opportunity for "My big fat greek superluminal experiment" jokes. You know, the one where the groom gets slapped by his future mother in law, then his future cousin in law tricks the groom into telling his mother in law that she has a nice pair of superluminal neutrinos... get it, she slaps him superluminally before he gets tricked?
at the moment they have merely found out that "data" sent over the fiber-optic cable arrives 60ns earlier then assumed
How does that happen? I've worked at fiber using telecom companies since 96 (customer and provider sites) and I've never heard of a loose cable causing 60 ns of constant delay. Random jitter as the connector bounces around? OK yeah. Intermittent loss? OK yeah.
You can trivially make a fiber "60 ns longer" but thats quite a length of extra fiber, not a tiny fraction of an inch.
My guess is someone thought they were purchasing a X yard long fiber cable, but the helpful installers put in a X meter long fiber without telling anyone, and the stereotypical telecom BS about loose connectors is the coverup for the situation. Or the gear is buggy, it stopped being buggy, and all the tech did was tighten the connectors, so "it must have been the connector". Uh huh, yeah, heard that one before.
In the book Schneier makes scores of astute observations on how society functions around security. He notes in chapter 16 that we are currently in a period of history where technology is changing faster than it ever has. The worry is that if technology changes too fast, the attackers will be able to innovate so much faster than society can
Historically haven't attackers always innovated slower at least on the net? I was on the net for years before the invention of spam around '93 or so. Most "attacks" seem to be the same old social con artist crimes, or finding dumb coding mistakes... but for those in the biz, those dumb mistakes are defined as dumb, not insightful. A buffer overflow is freaking magic to a noob, but to a guy who knows C its a parlor trick.
Technologically enhanced stupidity on the victim side seems to be a bigger issue than technologically enhanced criminality on the attacker side.
In 20 years on the net, I've seen the victims get stupider, but I haven't seen the attackers get smarter or dumber. You'd think the same demographic pressure would apply to each, but...
Dunbar gets made fun of sometimes because of the whole correlation/causation thing. Its just as sensible that social size and neocortex size depend on food pressure.
There are also the weird equivalents of the squared/cubed law
Also primate behavior is pretty boring compared to human, so Dunbar gets made fun of because of what amounts to FTE full time equivalent accounting. Surely the relationship I have with my cube neighbor is not exactly the same as yours, resulting in a larger or smaller chunk of the brain necessary to process... So 150 average full time equivalents, but someone really shallow might do the mentioned 3000 FB friends because each, on average, is only 1/20th of an average relationship. This is actually a well known FB problem, for example I don't do the 1/20th of a real relationship thing, so I found FB incredibly annoying, I don't give a F about some kid I sat next to in study hall 20 years ago, so I deleted my account from all the friendspam. A social media network 2.0 thingy that understands that would be interesting... G+ and its circle sliders is pretty close.
90% of the idiotic masses are going to call their ISP and scream at some poor script reader in India who probably knows nothing about this. 9% of the idiotic masses are going to call a fox news call in program and explain how its an indonesian commie plot to eliminate christianity from america, or some NPR radio show and ramble on about weed legalization would have prevented this in the first place and its all Bushes fault anyway. 1% of the idiotic masses are going to call 911 and they are gonna be pissed off
I'm guessing anyone smart enough to fix their box probably already has, so all you're going to get is hyper-flakey responses.
You don't "leave it on forever"/"shut it down forever". You turn it off from 0900 to 0901 today. Then 0900 to 0902 tomorrow. Then 0900 to 0903 the next day. Worst case scenario this BS is over in a mere 1440 days or about 4 years. Some people might freak out and fix it the first day, some people might not notice for a couple months, but eventually they'll all deal with it in their own way.
Option 4 which I guess outs me as a NANOG reader type of guy, is for an ISP or large corporation to BGP advertise the DNS servers specific netblocks as themselves (obviously route filter not to send to their upstreams or they'll get really pissed off) and run their own servers and then implement whatever they want whenever they want.
I don't do the windoze thing either at home or work, so I've been sorta ignoring this, but I think I read it was only 4 little/24s that need to get this treatment.
If you don't wanna run your own fake servers (well, technically they're just as real as the FBI ones...) then you just block those 4/24s in your firewall, perhaps temporarily, and "see what happens". Email goes out "if your internet was down from 1030 to 1100 today, please open a ticket with IT" etc.
Or your organization simply packet sniffs all traffic to the rogue server addresses and you follow up as appropriate. Obviously you sniff on the inside of the firewall, not outside, duh. This assumes you organization documents any of that internal stuff in a useful manner.
Slashdot Mirror
Slightly off topic, but very closely related, does anyone have stats on what results in more activity, structured scheduled programmed events or free time?
My gut level guess is free time results in much higher activity levels than structured events. Building a snow fort is exhausting compared to warming a bench at a sport. Etc.
It Might be that enforced structured scheduled wii play (OK governor katy pery song on just dance 3 will commence from precisely 1012 to 1015, after which the next scheduled song, like it or not, will be ...) is just boring and less energetic than kids just going crazy doing whatever they want.
Thinking back to my youth, when I wasn't chasing dinosaurs, hunting for woolly mammoth, or going fishing for trilobites, if I built a snow fort however I wanted I went crazy building it until I'd practically collapse, but given an "assembly line job" of repeatedly replicating a standardized boring data analysis driven snow fort I'd probably have all joy crushed out of the experience.
Short version is butchering wood in the basement workshop = big fun = high energy but middle school shop class making a birdhouse = no fun, I Fing hate birds = just do the minimum until next class. Wii might be the same way.
Anonymous and Wikileaks please help! Peoples lives depend on results kept hidden by two gangs of crooks, and only wikileaks and/or anonymous can save peoples lives... The government sure as heck won't help the people, the crooks are actively against us... the only people who can save lives are you two. go go go
1) One bunch of crooks makes money by letting people die in mines for higher profit. Evil personified.
2) One bunch of crooks makes money by charging both authors and readers to distribute research funded by taxpayer dollars. Pointless dying intermediaries.
I would love it if both evil groups of crooks get screwed. Anyone got a link to a torrent on the pirate bay yet? Or a .onion tor link, or a i2p link, or ... note I'm not looking for links to rickrolls and goatse, only links to the genuine documents (assuming there's no false flag operation where the mining co. themselves release "massaged" data, not outside the realm of possibility)
Are there no superheros anymore? Is this just too difficult for anon and wikileaks? Come on guys, get it in gear and save some lives.
Windows is familiar, Linux isn't.
Windows used to be familiar. The difference between the last decade or so of windows UIs is ... immense
a quick (bash runonallmachines.sh apt-get install foobar) could print out the info to stdout
This seems to be turning into an endless array of "lets find even harder wheels to buggily reinvent". Its NIH all the way down each step of the way... The proper way to do this wrong thing in this situation, is to "apt-get install dish" which is the distributed shell package. Put yer list of hostnames in ~/.dish/hosts and yer options in ~/.dish/options (-p0 and -T 300 are fairly popular...). Any sane person has already figured out ssh keys and ssh agent and sudo all that before hand. Then its just "dish apt-get install foobar"
Also if you've ever tried this unique form of cluster package management, its a fail if apt-get blocks on some debconf questions, or blocks on waiting on input asking if its ok to install dependencies. Also fails if any previous package installation failed awaiting dpkg-reconfigure -a or whatever. And it fails if something went weird during the most recent (perhaps automatic?) apt-get update. Again, you've not figured out how to manually process the logs or manually process failures.
It is true that there exists a way to do package management which is harder and slower and more tedious, and that is to manually by hand ssh into each machine and run each command by hand. But finding a solution one millimeter above the absolute bottom of the barrel, doesn't mean you've found the optimum solution.
You are correct, in that you've just found a way to avoid the staggering and overwhelming job of adding this line to some init.pp file on ye olde puppetmaster:
package { "foobar": ensure => latest }
Seriously, using puppet really IS that hard. We're not talking about writing a Turing machine in sendmail M4 macros here. Puppet was designed to do Exactly what you're trying to do, and do it very well, very easily, and very quickly. So, just do it... There's absolutely nothing wrong with hacks for the sake of doing a hack, or reinventing something to figure out how its done. But when its time to work, use the right tool for the job. Don't install drywall screws with a hammer.
it became cheaper and easier to locate decent beer so people that brewed just to get good beer no longer needed to brew at home
That's like trying to argue that people only cook at home because there are not enough restaurants. "If they'd just open a Thai restaurant around here, then architects could stop putting kitchens in homes". Don't think so...
I'll be honest, several of my experiments in brewing tasted awful, much like some of my cooking experiments would have been best not eaten. The fun is in the experience of making it myself, my way. Its like solving a big puzzle.
The other part is brewing is "big enough work" that it becomes a social activity. You can make a social activity out of the alternative of driving to the liquor store and waiting in the checkout line, but trust me that brewing is much more fun.
The biggest problem I have with microbrews at the store is they sit there long enough under the lights to get old and skunky. Might have been good fresh, not so good after 6 months getting dusty on a shelf. Yuck. Laughably some people think skunky beer is the "new taste" microbrews are aiming for...
I believe beer is the perfect lens through which to examine innovation,
Here's the most important lesson which I bet is either not covered accidentally or maybe intentionally.
I live in a greater-city which used to be the center of American beer brewing. A century or so ago, German immigrants built dozens of medium sized breweries and exported all over the country. Big big names, still around in marketing even today.
All of those jobs, and I mean all, are gone, inside the city. Every last one. Mergers inside the country and international, centralization, downsizing, blah blah, and now we've gone from dozens of breweries to a handful of microbrews, depending on how you want to count Sprecher (in a nearby city) and this brewpub by the local engineering college. A century ago there were dozens of people in my city with the job title "brewmaster" now there is debate but the number seems to hover right around "one" or "zero" depending how picky you want to be.
Similar thing happened in the automotive business, from hundreds of companies a bit over a century ago to just a handful now. Same deal multiple times with computing.
The lesson is that in a Emerging Technology there might be thousands of management and engineering jobs, but eventually its no longer an Emerging Technology then almost ALL of those jobs go away, permanently. If you're a 1 in a 100, maybe you can be a survivor making a long term career out of emerging tech, or if you enjoy perma-unemployment after a real fun 10 year run that'll work, but otherwise, if you see emerging tech, run like hell away, if you care about your family being able to eat and have a roof over their head. Run!
I've heard it said the difference between a homebrewer and a pro is like the difference between a internal skeleton and an exoskeleton, in that a homebrewer sanitizes by putting stuff into the kitchen (or basement) sink, whereas the pro takes sanitizing solution out of the sink and into the apparatus because its so big.
I sell home beer and wine making supplies and ingredients in my hardware store.
As a guy who brewed beer in the past and probably will again in the future, most of the stuff you sell to brewers, you probably don't know about.
I used to buy replacement plastic transfer hoses, copper tubing and handful of strange compression fitting adapters to make my own wort chiller, tubes and hoses to make my own homemade bubbler/vaporlock thingy, etc etc. I purchased all the gear to make what amounts to a remote faucet system on a hose for cleaning. I had the worlds weirdest rube goldberg device to fill bottles. For wine/mead supposedly the most expensive and traditional primary fermenter is a glass carboy, and supposedly the cheapest is a food grade plastic bag (not insecticide treated garbage bag) inside a non-value engineered old fashioned strong metal trash can. Supposedly prices have exploded upward so much that the cheapest durable and watertight primary "couple gallon" fermenter is a standard tropical fish aquarium, although keeping light out and the top sealed must be a huge PITA.
I never bought "normal homebrewing stuff" from a hardware store like yeast and hops, bottle caps for my crimper, whatever. Thats cool that you sell that stuff as I have 4 hardware stores within 5 miles, but my local "homebrew store" was at least an hours drive. In the internet era its more realistic to order online and wait a day or two, than to invest that kind of windshield time.
Speaking of which, can you even buy non-cordless phones these days?
You have to go to walmart, but they do exist.
The real problem is you need a copper line that is homerun to a CO not to a SLC hut. SLC hut battery backup is ... not so good, if there is any at all. Supposedly there existed a SLC-96 system 30 years ago that was CO powered off the T-1 repeater supply, but they've only installed fiber SLCs for decades now and the few I've seen the insides of have metered electrical service.
People who only know a little telecom think every copper line is a home run to the CO. People who worked in telecom know that 20 years ago SLC market penetration was at least 1 in 10 residential lines, and now, I would not be surprised if the majority of copper lines are run to a SLC.
It does depend on your neighborhood. If there's a homeless panhandler on the sidewalk, thats urban and probably 100 years old and you probably have copper homerun, unless its a "factory to condo conversion" and the telco put a SLC in. If its a "1950s baby boomer house" like mine then its 50/50 and in fact mine is on a SLC (which was a nightmare to get DSL back in ye olden DSL era). If its a modern mcmansion I guarantee the LEC installed a SLC, they're not going to home run copper all they way from each house to the CO.
LEDs are used to cycling at 120 hz, or I guess 100 Hz in soccer-hooligan-land. This might be an excellent excuse to upgrade to LEDs.
Does anyone in the biz know what they use to drive the LEDs? I'm assuming a simple bridge rectifier and some manner of constant current switcher, but "real genuine streetlights" might do something more exotic, I donno.
Note I'm not interested in how model railroader or RC car builders bias a LED using a simple resistor, I'm quite well informed on that, thanks, I'm looking for real details how real streetlight ckts work.
Of course, the control system required is far more complicated here. I wonder how much energy is consumed in producing and maintaining the new lampposts, controls, communication network, etc.
I put on my engineers estimating hat and...
I will call 1 KW per lamp. Yes its probably beneath but not much.
Cost per KWh is below 10 cents/KWh for everybody, far below for a major account like my city or the highway dept, but its "In the ballpark" when you add in corruption, management, accounting, overhead in general. So 1 KW lamp costs $0.10 per hour
yearly average is 12 hours per day of lamp and 365 days per year but thats a pain to multiply so we'll call it 10 hours a day and 333.333 or a 1/3 of a thousand days per year, for a third of a thousand of ten tenths bucks per year or simplifying a third of a thousand bucks per year.
I don't see why they'd have to replace the lampposts. The control gear is not going to last 25 years or whatever mechanical timer relay based controls lasted, this is going to be a MS windows installation with no upgrade path other than forklift an entirely new system in, so lets call it optimistically 10 years.
So over the life of the install, you would spend $3333 for a dumb system, so your smart system cannot physically save more than $3K or so, so unless its the usual greenwashing corruption it needs to cost an order of magnitude less, like a couple hundred per pole absolute maximum.
Which, looking at X10 / insteon type stuff is doable, but if you want to use real scada controlled PLCs is probably not doable. It'll be a tough installation.
Also before anyone pipes up, I'm sure the output changes will increase wear and tear on the bulbs, but thats counterbalanced by the cost of electricity increasing over time, so I think it fair to disregard.
One of the problems about dimming lanterns is that the lamp post spacing is all based around the lamps at a certain luminenscence and so dimming may create dark zones, or over bright zones. So some careful analysis will be needed about how the lamps dim and whether they dim uniformly or not.
I can personally verify this zone problem happens with at least some LED lights. I live in a city that has been testing old fashioned LED lights (a couple years, maybe a decade behind China at least) and they're trying to retrofit the LED heads onto existing poles but the heads they chose do not output the same light cones as the old sodium vapor or whatever lamps. So we get bright and dark bands flickering as we drive. I am not epileptic but I know the flashing sets them off and its bad enough to be disturbing to me, so it must be absolutely awful for them. Also the glare from the LED point sources is intensely dazzling to the point of danger whereas the older lamps did not have as high of a peak glare output.
I live in snow country, and the LED streetlamps do get hot enough to melt the snow off them. Must be a hundred watts of luxeon up there. The very well known problem of stop and go lights completely encased in snow and ice does not happen with the LED streetlamps, or at least not the ones we have.
This does not absolutely prove LEDs suck; its entirely possible the laborers who made a living replacing old fashioned light bulbs convinced someone to intentionally install a batch of fail for the test program so they can keep their jobs after we rip out the horrid LEDs.
Intel AppUp Center is an online repository designed for Intel processor-based devices.
Minor correction; its a windows only app store. Does not perform the miracle of running the same executable on mac osx, all linux distros, and windows. Just windows thats all.
libreoffice is available for all those platforms, just not available on the windows only appup
If you only have 30 machines for one event, puppet/cfengine is overboard. Just set up a passwordless SSH key for root (remember NOT to put the private key on the laptops), and just use a simple script to send the same commands to every laptop.
You forgot error recovery and logging. So PC #25 was rebooting while your script ran... does that make the script fail, perhaps silently? Does that mean all the PCs except #25 are OK, or #25-(end of list) fail? How do you know to rerun the script later? How many times do you have to run it by hand to "make sure" it ran on all machines? You can add logging and some sort of retry mechanism... Just remember that those who try not to use puppet, end up rewriting puppet, just takes a long time and painful bugs. apt-get install puppetmaster on the main and apt-get install puppet on the remotes is just too easy to bother writing a clone of puppet out of shell scripts.
Puppet is much like AFS or LDAP or well, practically any service, the first time you set it up you're all "WTF?" and the second time its no big deal. Google for some tutorials, screencasts... Doing a really halfway job writing it yourself is terrifyingly harder than just using puppet.
http://puppetlabs.com/
LOL this is the weekly ask /. where the questioner describes the perfect application for Puppet and then asks what to use.
My only other addition is install and set up torque and dish.
Torque is a decent queuing system. Everybody queue up a job to do "something" as quickly as possible, but strictly one at a time.
The DISH distributed shell lets you run a single command line (which could be a script...) on all machines right now. Simultaneously or whatever.
With all due respect, nothing personal, but the ideas you expressed are completely wrong. Kids need to learn that science is experimenting and debating and arguing and trying things that mostly don't work but sometimes they do. There is no cabal and smart people sometimes disagree, most importantly they disagree in a civilized manner. And getting excited and theorizing and double checking your work and then triple checking your work and lots of sweat and effort and long hours. Initial results are sometimes wrong. Where do errors come from? And sometimes how you deal with "failure" defines who you are, more than how you deal with "success".
Science is not (or should not be) a scholastic endeavor that we should try to make as boring and authoritarian and slow and uninteresting as possible. If anything try to make it the opposite, at least a little bit.
If this whole story makes one kid think, just a little bit, about physics, that makes it OK. This is the best thing thats happened to physics in years.
If science were as flaky as a reality TV show, then I'd support your position because somewhere in between is the greek ideal. But... there's a long way to go before we have to worry about that.
I have to think about this for a second, but I'd assume the neutrinos had a Very accurately measured speed. So maybe the story isn't that the neutrinos go supraluminal, its that fiber optic cable at CERN has a slightly negative velocity factor...
Now you'd think people have measured the length of fiber with a OTDR before (god knows I have enough time) but maybe there is something weird about CERNs fibre, like they had to wrap the slack somewhere and they had a 10 tesla superconductive magnet laying around of a convenient diameter, so... , and no one has ever OTDRd something that stupid before (although I remember doing some stupid OTDR tricks when I was learning in the 90s)
>Fuck it, I'm going to Greece.
>Go to Greece.
>Considering marrying a Greek girl
>Berlusconi is there
You missed a golden opportunity for "My big fat greek superluminal experiment" jokes. You know, the one where the groom gets slapped by his future mother in law, then his future cousin in law tricks the groom into telling his mother in law that she has a nice pair of superluminal neutrinos... get it, she slaps him superluminally before he gets tricked?
at the moment they have merely found out that "data" sent over the fiber-optic cable arrives 60ns earlier then assumed
How does that happen? I've worked at fiber using telecom companies since 96 (customer and provider sites) and I've never heard of a loose cable causing 60 ns of constant delay. Random jitter as the connector bounces around? OK yeah. Intermittent loss? OK yeah.
You can trivially make a fiber "60 ns longer" but thats quite a length of extra fiber, not a tiny fraction of an inch.
My guess is someone thought they were purchasing a X yard long fiber cable, but the helpful installers put in a X meter long fiber without telling anyone, and the stereotypical telecom BS about loose connectors is the coverup for the situation. Or the gear is buggy, it stopped being buggy, and all the tech did was tighten the connectors, so "it must have been the connector". Uh huh, yeah, heard that one before.
In the book Schneier makes scores of astute observations on how society functions around security. He notes in chapter 16 that we are currently in a period of history where technology is changing faster than it ever has. The worry is that if technology changes too fast, the attackers will be able to innovate so much faster than society can
Historically haven't attackers always innovated slower at least on the net? I was on the net for years before the invention of spam around '93 or so. Most "attacks" seem to be the same old social con artist crimes, or finding dumb coding mistakes... but for those in the biz, those dumb mistakes are defined as dumb, not insightful. A buffer overflow is freaking magic to a noob, but to a guy who knows C its a parlor trick.
Technologically enhanced stupidity on the victim side seems to be a bigger issue than technologically enhanced criminality on the attacker side.
In 20 years on the net, I've seen the victims get stupider, but I haven't seen the attackers get smarter or dumber. You'd think the same demographic pressure would apply to each, but...
perverted religion teaches deceptions, lies and murder
Yeah, that really narrows it down to ... all religions (except Buddhism). Oh wait I see you mentioned one by name in the first line. Sry.
Dunbar gets made fun of sometimes because of the whole correlation/causation thing. Its just as sensible that social size and neocortex size depend on food pressure.
There are also the weird equivalents of the squared/cubed law
Also primate behavior is pretty boring compared to human, so Dunbar gets made fun of because of what amounts to FTE full time equivalent accounting. Surely the relationship I have with my cube neighbor is not exactly the same as yours, resulting in a larger or smaller chunk of the brain necessary to process... So 150 average full time equivalents, but someone really shallow might do the mentioned 3000 FB friends because each, on average, is only 1/20th of an average relationship. This is actually a well known FB problem, for example I don't do the 1/20th of a real relationship thing, so I found FB incredibly annoying, I don't give a F about some kid I sat next to in study hall 20 years ago, so I deleted my account from all the friendspam. A social media network 2.0 thingy that understands that would be interesting... G+ and its circle sliders is pretty close.
90% of the idiotic masses are going to call their ISP and scream at some poor script reader in India who probably knows nothing about this.
9% of the idiotic masses are going to call a fox news call in program and explain how its an indonesian commie plot to eliminate christianity from america, or some NPR radio show and ramble on about weed legalization would have prevented this in the first place and its all Bushes fault anyway.
1% of the idiotic masses are going to call 911 and they are gonna be pissed off
I'm guessing anyone smart enough to fix their box probably already has, so all you're going to get is hyper-flakey responses.
Hooray for /. binary thinking.
You don't "leave it on forever"/"shut it down forever". You turn it off from 0900 to 0901 today. Then 0900 to 0902 tomorrow. Then 0900 to 0903 the next day. Worst case scenario this BS is over in a mere 1440 days or about 4 years. Some people might freak out and fix it the first day, some people might not notice for a couple months, but eventually they'll all deal with it in their own way.
Option 4 which I guess outs me as a NANOG reader type of guy, is for an ISP or large corporation to BGP advertise the DNS servers specific netblocks as themselves (obviously route filter not to send to their upstreams or they'll get really pissed off) and run their own servers and then implement whatever they want whenever they want.
I don't do the windoze thing either at home or work, so I've been sorta ignoring this, but I think I read it was only 4 little /24s that need to get this treatment.
If you don't wanna run your own fake servers (well, technically they're just as real as the FBI ones...) then you just block those 4 /24s in your firewall, perhaps temporarily, and "see what happens". Email goes out "if your internet was down from 1030 to 1100 today, please open a ticket with IT" etc.
Or your organization simply packet sniffs all traffic to the rogue server addresses and you follow up as appropriate. Obviously you sniff on the inside of the firewall, not outside, duh. This assumes you organization documents any of that internal stuff in a useful manner.