Slashdot Mirror
The Grid, Our Cars, and the Net
Wired is running a piece on the big idea of Robin Chase — the founder of Zipcar — that we need to build our smart power grid on open standards and include cars as nodes in a mesh network. "'Today in Iraq and Afghanistan, soldiers and tanks and airplanes are running around using mesh networks,' said Chase. 'It works, it's secure, it's robust. If a node or device disappears, the network just reroutes the data.' And, perhaps most important, it's in motion. ... Build a smart electrical grid that uses Internet protocols and puts a mesh network device in every structure that has an electric meter. Sweep out the half dozen networks in our cars and replace them with an open, Internet-based platform. Add a mesh router. A nationwide mesh cloud will form, linking vehicles that can connect with one another and with the rest of the network. It's cooperative gain gone national, gone mobile, gone open."
no longer will we be slaves to the ISPs!
Does this mean I can be alerted when I get near someone who cut me off before?
Imagine it, man. It would be like so many little nets inside of bigger nets going on forever (deep inhale, coughing exhale). Wow, uh, it would be like TRON, only way better. I need a salty snack.
And the crackers and vandals will pee themselves with joy.
Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
Big ISPs and phone companies have too much to lose to allow this to ever happen.
It would be too hard to be tapped by various 3 letter government agencies so they wouldn't like it either.
Maybe instead of continuing to focus on the dinosaur that is the automobile, more effort should be put into building very a efficient mass transit infrastructure. Just a thought.
Hi there
What we need is to have it integrated into our phones and that we can tether to so:
a) consumers choose phones with over phones without
b) we can use it even outside the car and
c) it's not connected to cars (better to stop the car rebellion right there, tyvm).
A nation-wide cloud? That sounds pretty bad. I heard the fumes were toxic, that's why I stopped making it in my basement. ... OH! MeSh cloud. I see.
I vote thanks but no thanks on this. Despite whatever wild-eyed claims about "openness" or "oneness" or whatever other hippie bullshit the brainchilds of this are spouting, there is absolutely NO information of any kind that is appropriate for my vehicle to be broadcasting. I'm sure the police and Federal government would absolutely LOVE to have a way to track the location of every vehicle in the country, not to mention who owns it and who they're talking to via their built in net cellphone at the time. Integrating this with the idea of a vehicle is a hilariously bad idea, because the instant it comes about there will be DOT, Federal, and State laws with a laundry list of mandates about how "open" this system will be allowed to be to be "roadworthy," and I guarantee you not a single one of these mandates will be in your best interest.
Pass.
If we're going to do the mesh network thing, I'd rather have it in a portable device like a phone or PDA that doesn't give the government a billion inroads to regulate, legislate, and subvert it, and one that I can chose not to buy, to turn off, or to leave at home.
I can't say I completely understand the article. It seems either she, or the person who wrote the article, is confusing mesh networking with power distribution. The article doesn't make clear how the two fit together (maybe someone else who understood can explain better). It talks about wireless networking at the same time it talks about plugging things in. Those two don't seem to fit well together (yeah, I know, some companies are developing wireless electric device chargers, but it's a totally different concept).
One thing that interested me in the article was this quote, " the Obama Administration allocated $4.5 billion in the stimulus bill for smart grid R&D." So we're getting some kind of smarter grid anyway, at least some research into it.
Qxe4
15 years ago when i looked at the literature there were substantial problems with the efficiency of the selected routes, route convergence and message overhead. these things got much worse as the rate of change in the peer graph goes up.
have things gotten that much better?
... will plugging my car into this "mesh" gain me? I don't see a reason for this. It's excessive and prone to more problems than we already have (I guess. I don't even understand exactly what problem she's trying to solve so as to properly determine that). I don't see the automobile in the same light that she does. Just let my car be a car and be powered by my power, Mrs. Xzibit.
The eternal struggle of good vs. evil begins within one's self.
Or something like that.
Those who actually bothered to read TFA, what exactly is the point of this? I understand Robin Chase loves feel-good social causes, and she is a good organizer, but no one ever accused her of being an engineer. Having read TFA, it sounds to me a bit like confused meandering of someone trying to figure out how to use some of the stimulus billions for yet another social pet cause, but without the clear definition of what that cause is.
End anonymous moderation and posting on
The CIA is obviously using a very loose definition of the term "urban". They're probably counting suburbs around large cities as being IN the city.
"The tree of liberty must be refreshed from time to time with the blood of patriots and tyrants." ~Thomas Jefferson
So once this is done, can I IM other drivers and give them driving tips? Like, oh, I don't know -- "Hey buddy! If you'd bothered to use your freakin' turn signal, I coulda made my left, and not had to wait another ten minutes for an opening in traffic! Thanks a lot!" Because, if so? Sign me up!
I am not left-handed, either!
My point isn't that mass transit should be ignored, or that we shouldn't look at doing it in a more effective way where it makes sense, but that wide scale mass transit that has as a stated goal of replacing the automobile in most circumstances, even in rural America, would not be advisable. There are about 303 people per square mile in France (non-euro territories not included), compared to 33 people per square mile in the US (excluding Alaska's area). I've excluded the no-mans-land of Alaska from this equation, but even if you excluded all of the areas where nobody lives in America, you'd still have a significant density difference between France and the US. The issue is not the same here as it is there by a large margin.
The other things is, I was in Pittsburgh in the early 90's during the Port Authority strike - no busses or trains ran in Pittsburgh for a week or more. What I recall is that people found ways to get wherever they needed to go, the air was SIGNIFICANTLY clearer and cleaner without all of the diesel belching busses on the road, and even though everyone had to get to work by private vehicle, traffic moved BETTER, because the slow-assed busses weren't clogging traffic up at every intersection in the city during rush hour.
Folks who write articles about smart grid communicating with cars, etc bring to mind foolish talk of internet toasters and networked refrigerators.
The current electrical grid (speaking of USA; PJM region in particular) is very reliable as it is. Grid operators already have the ability to shape production; with millions of users, usage patterns are easy to spot and plan for ahead of time.
In my view, smart grid and smart meters are simply a way to control people's usage and charging them more; residential electric bills will likely become very complicated.
All this talk about people charging their cars at night and then selling it back during the day for extra credit is nonsense, because when millions of people are charging at night, it's easily conceivable that nightly usage could be just as high as during the day.
In respect to cars communicating with other cars - why? It's obvious that most people will charge their cars as often as possible, even if told not to, in particular, at night so they are sure to have enough charge to get to get the kids to school, get to work, etc.
The internet is another means of communication - it's not going magically solve energy issues nor change human nature.
In my view, a better approach than a so-called smart grid is developing / promoting more efficient energy production methods, in particular nuclear (solar, wind, etc are nice, but are lacking in energy density), along with encouraging people to produce some of their own energy for their needs, such as with solar panels on their roof.
Ron
Comment removed based on user account deletion
someone trying to figure out how to use some of the stimulus billions for yet another social pet cause
Bingo.
You will see more of this soon.
Hello? Sorry, I'm going to be late for work. I'm trying to get out of my driveway, but the car just says "Buffering..."
http://twitter.com/OLDTELEGRAM
So, the internet could someday actually *be* a big truck?
Having read TFA, it sounds to me a bit like confused meandering of someone trying to figure out how to use some of the stimulus billions for yet another social pet cause, but without the clear definition of what that cause is.
My feelings exactly. It has lots of woo-woo words and ideas which seem magical and yet, I can't understand what the fundamental idea is exactly. --It almost sounds like she's suggesting that we use phone system-like switching technology to route power to individual homes and devices. Sounds bloody expensive to put into place. A high voltage router on every street corner, though I don't really see the advantage, unless each house is also generating electricity. Maybe I'm missing something.
If the concept is viable, then it can be explained in baby terms, which clearly I need. I feel like one of those really thick studio execs guys like Kevin Smith make fun of. So either I'm really, really dumb, the idea needs work, or she needs a good translator to stand between her head and the audience.
Those Zip Cars looks sort of cool, though I don't quite see the advantage. Do they have a team of service people running around each city maintaining the cars? It sounds like a de-centralized "Rent-a-Car", but while they don't have to rent a main lot, they still must have to maintain a garage and offices somewhere, and I bet they have to pay for all the individual parking spaces. Seems gimmicky, but again, maybe I'm just not seeing the big picture. (After all, I still refuse to use a cell phone over my trusty land line, so it's entirely possible that I'm missing the point.)
-FL
I did a lot of tariff programming back in the day and I loved it...
Electrical demand is not the same as network demand. If your ISP is short on bandwidth, everyone just slows down. But if your power company is short on power, at worst, they have to start throwing people off of the grid, because everyone must have 110VAC 60hz power.
This reality is reflected in the pricing of electricity, especially for larger customers.
The kind of an electric bill a refinery gets, for example, shows this. In such bills, you start with the raw data obtained from power recorders - every kwh and kvarh (reactive power), is recorded at either 15 minute or hourly increments, depending on the utility. This data is rolled up to look at peak demand, and bill to date usage, broken out into buckets representing time of use, each of which has its own price. For the most part, the demand portion of the bill is roughly half, and the other half is the cumulative portion.
So, of all the actionable items in a bill that one could act on, really, instantaneous demand is the most important thing to optimize. If you jack up your demand during the day, for just one hour, by 50%, you've significantly increased your monthly bill... because the utility still has to have equipment to satisfy peak service.
The thing is, industrial customers have known this now for at least 10 years, if not longer, and there's a whole electrical services industry designed to help them avoid that maximum demand charge. Some companies making ice at night for cooling by day. Others try and have multiple shifts. Still others just put in their own local generation that kicks in when their utility usage gets too high. All of this is controlled by automated SCADA systems that have been field proven for at least a decade, if not longer.
The point is, I'm wondering how much smarter the electrical world can actually get? What you are really talking about is putting residential customers on industrial style tariffs. But, what would be the benefit? I mean, there's not too much a residential customer could practically do that would cost effectively help them lower their peak demand in such a way as to be cost effective.
For example, in California, for SCE, the GS-2 tariff specifies a demand charge of less than $10 / kw. SCE GS2. If you figure that most homes use less than 2Kw max demand, there's not much room for economical demand shaving. If you lowered your peak demand from 3kw to 2kw, you would be saving $120 a year. There's few, if any devices that could store energy at night, help with peak demand by day, where you could actually recoup that investment economically.
This is my sig.
The mesh networks exists. Take a bunch of wifi routers, mesh their ssid's together. I imagine this to be a mesh network. No surprises.
The current WIFI users are slow-moving at home and only using WIFI at home.
The "TWIST" in Ms. Chase's opinion is to design WIFI for FAST-MOVING-WIFI users. Design wifi for moving quickly in and moving quickly out of a particular user's range of acceptable signal strength for reliable connecting/sending/receiving data.
The popular network apps like firefox/bittorrent still assume wired networks. That will change with time however. I'm sure a kind of wififox or wiftorrent designed for "UDP protocol" already exist for 3G/2G phones using proprietary source code. That's Ms. Chase's point. If that code were made to be open-source code, the world would greatly benefit. She didn't say anything about making the node-hardware easy to adapt to an owner's needs, but I think she was implying it by bringing up the open-source approach. Open-source implies "DIGITAL FREEDOM" as www.fsf.org/campaigns would express it. Every FAST-MOVING-WIFI-USER should have the ability to modify one's FAST-MOVING-WIFI-HARDWARE as one's sees fit.
I'm absolutely sure the guys that built UDP a long time ago already did everything necessary for mesh networks.
For as long as your computer is connected to two networks simultaneously. If the first try fails then you can retry the second packet on a second/third/fourth... network simultaneously. The first response back wins.
All these wifi devices already have a unique ssid manufactured into them. Right now the ip version 4 address and ip version 6 address needs to be changed if we moved from one wifi router to another. Reliable TCP/IP v4/v6 communication assumes your IP address doesn't change under your feet every second. I.e. SSL assumes the same constant source IP address and the same constant destination IP address for the connection to stay up.
But there's UDP...When using this protocol, it is understood and implied that the connection will not always be up. It is understood the protocol needs to be prepared for unreliable packet communication. This makes UDP more suitable for for wifi. The source ip address needs to be acquired from the wifi router. ok. the source ip can send a UDP request packet to the router and pray that the router will be quick enough give him a response UDP packet back. If he doesn't then it's time to send out another dhclient client request to get a new IP address from another router. Then reconnect to whatever other node you're talking to and continue to send/receive whatever. The routers know UDP by the way so it's just a matter building the applications with more wifi context.
Are there such api functions as:
getIpAddressForSSID(SSID as string)
returns IpAddress as string
getIpAddressForMAC(MAC as string)
returns IpAddress as string
?
They would be useful for the UDP/FAST-MOVING-WIFI.
My guess is that they exist, but I've never had a requirement to use them myself.
You can't just slap a mesh router on a car and expect to be able to pull up
Mesh networks are great in some situations, but not in vehicular networks on a city-wide (or country-wide) scale. 200 mesh routers in Vienna is cool, but what about 500,000, moving at an average speed of, say, 30mph? In this case:
Cooperative gain means more users bring more capacity, not less.
Is absolutely not true unless you're talking about storage volume rather than bandwidth. I can't wait to see someone spend millions of dollars to equip 10,000+ cars and watch everything fail, as long as they don't take away from my research funding.
TLDR: Leave the mobile networking ideas to the people who know what works and what doesn't.
She read some academic papers on mesh & vehicle networking (DARPA funded for sure), and pooped.
I hate to break it to you, but this far everything everyone's ever penned on this particular subject because... are you following along with this at home? The technology doesn't exist yet. We don't know. It's an unknown, and anything could happen up to and including nothing at all.
All I'm doing is putting forth a possibility tainted by my own opinion. I'm straightforward enough to be honest about that. Why are you so hell bent on turning speculative opinion into some kind of bullshit personal affront?
While we're on that topic, though, let's look at the track record of our illustrious elected officials. It is already mandated that all cell phones sold in the US come with GPS chipsets capable of transmitting your phone's location. In most cases this can be disabled excepting 911 calls, but the technology is there. Your position can already be triangulated fairly accurately just via cell towers. This is a proven fact. Warrantless wiretapping and general spying on US citizens without cause by the Federal government is so well documented that there's already been massive outcry and a million and one headlines about it. This is a proven fact. OBD2 compliant automobiles sold in the USA are required by the DOT to have black boxes (for lack of a better term) that record vehicle speed, brake status, RPM's, and the other assortment of telemetry available in a modern engine for the sole purpose of the police using it against you in post-crash investigations. This is a proven fact. Traffic cameras are already in place in many locations throughout the country and are not only used to hand out speeding tickets as well as track individual vehicle movement when the police so desire, as has made headlines more than once. This is a proven fact.
None of the above is speculation. People who live off your tax dollars want to know where you are and what you're doing at all times, and the demonstration of this desire is made clear again and again and again. How many stories are posted to Slashdot to the effect of "company developing X technology to recognize faces/scan fingerprints/track crowd movements/snoop on cell or internet conversations?" Count them. How many of them go on to say they're doing it with government funding or for homeland security purposes, and all those other buzzwords? There's a reason Slashdot has a "Your Rights Online" section. There's a reason stories like these are of so much interest.
What is speculation is what will happen if a widespread vehicle based network of no concrete design or aim is put into service. My speculation is that bad things will happen if it is, especially given the track record of the US government both Federal and local in passing mandates involving automotive technology. If you don't agree with my speculation, that's fine. But if you want to blow it out of proportion and turn it into some kind of affront that's all you.
As soon as someone claims something is 100% secure I know they don't know what they are talking about.
I'd rather walk.
People who live off your tax dollars want to know where you are and what you're doing at all times,
Oh boy, believe me, there are times that they'd probably rather not know what I'm doing.
In addition, I am sure that my doings are well under the radar, so to speak.
Karnal
Diesel has nothing to do with it; diesel is up to 40% more efficient by the gallon than regular gasoline (although it produces ~10-15% more greenhouse gasses). The reason you see dump trucks and buses belching black smoke is because they're so ridiculously heavy and they have to burn a lot of fuel to get going. Anyway, my city uses natural gas or something to fuel the buses. Also I agree with you that obviously mass transit is not an option in rural areas but pollution isn't bad in rural areas; we really only need mass transit in urban areas, and we need it adopted by everyone.
I've always thought that gridlock is partly a product of all the micro-pauses between seeing the car in front of us move and reacting to that. A grid system might be able to eliminate these pauses and hence remove what seems to me, a major cause of congestion. A grid might also be able to help with other similar traffic bottlenecks, such as merging and changes in direction. The only problem is, you'd have to give up some control of your car when in heavy traffic - but hey, you never had much anyway.
If you're challenging my assertion that there was a noticeable improvement in air quality in Pittsburgh when the diesel busses were not on the road due to the driver strike, you're wrong. Diesels produce up to 400 times the amount of particulate pollution vs. gasoline engines. Mass transit busses remain on the roads all day and most of the night, unlike a commuting auto, which is normally used an hour or two per day. There was a significant improvement in air quality when those diesel busses were not running - it felt literally like a breath of fresh air.
Busses running on natural gas would be a big improvement for urban areas, and it appears natural gas is being used more frequently now.
As for needing it adopted by everyone, you go right ahead. The less time I have to ride around in a closed enviro with every kind of virus carrying, shower-phobic, alcoholic, tourette syndrome suffering mental case in town, the better.
Buses carry more than one person per hour or two.
I think this is an excellent idea, so long as there are limits to it. For instance, the network should NOT be connected to, or capable of connecting to, the actual functioning of the vehicle. Worst case example being it should NOT be even vaguely possible for someone to hack into your car and turn off the engine. Or slam on the brake/gas. That being said, being able to read/broadcast status reports would be good. Such as letting the driver behind me know that I just slammed on the brakes. I know tail lights are supposed to do that, but anyway. Or possibly acting like the yellow flag at the racetrack : "a car 500m ahead just lost control, be careful". Likewise, for areas subject to snow storms, fog, or other conditions of poor visibility, such tracking would be *very* appreciated just knowing how far away the next car is in front/behind. Likewise, if such a mesh network was actually part of the internet, it could conceivably make it possible to connect to the internet, access it, and whatnot without going through *any* ISP. On the one hand, the potential for tracking should worry the privacy and rights advocates, and with good cause. On the other, such distribution of networking could also enable rights and privacy, since it'd be hard to track anything through such a constantly changing network, and even harder to filter anything. I would say it pretty much eliminates the option of filtering our internet. All that being said, I'm not too sure what the connection is with the electrical grid, aside from the potential of "smart" use of electricity, which would be good.
Z
my guess is that zipcar probably pays alot to provide the network that their cars currently use.
Yeah, I really need to find a new shtick, don't I?
This just in... No, wait.
Allow high load devices to communicate with energy producers to reduce peak load on the grid by rescheduling periods of high load.
Well, I managed 25 words or fewer. Basically, electric cars will all get plugged in when everybody gets home from work. If they all try to charge as quickly as possible, you get an extreme load on the electric grid from ~6-9 PM. Most of those cars won't be driven again until after sunup when the driver has to get to work the next day. It would be more efficient if the cars charged from 6pm - 6 am. If the power plant can tell the cars "hey, I'm going as hard as I can. Try later tonight when everybody is asleep!" then the electricity distribution system can be built around a typical average load, instead of needing to be built to handle wasteful peaks.
Then, add lots of hippie buzzwords about interconnectedness and unity. And some businessy buzzwords about paradigms and efficiencies of marketitude.
There's an open source EV project called the Tumanako Platform - take a look at what's going on: http://www.tumanako.net/
Sweep out the half dozen networks in our cars and replace them with an open, Internet-based platform.
Most vehicles have just two networks, one for components inside the cabin and one for the rest of the vehicle. Both are implemented with hardware that's way too primitive to cope with IP. They're completely datagram-based services whose primary requirement is to be able to cope with a large amount of line noise, hence the protocols are slow (typically 250Kb/s or thereabouts). Opening these networks up to internet access gains nothing, except perhaps the ability for a hacker to remotely disable your car while you're driving it.
Try another way (today) to move 3 tons of steel, etc, 12 miles down the road for less than $2.50. Plus you have the option of towing an additional 6+ tons for maybe an additional $1 every 12 miles.
This is not to say we can't imagine more efficient ways. The trouble is not inefficiency, but the use of limited resources (fossil fuel) and the stuff coming out of the tailpipe.
And why shouldn't those suburbs be included in any decent mass transit system? Isn't that where most people live who go to work in the city every day?
A very good (old school) friend of mine had a run in with some blackmailers a while back - the whole "give us money or we'll damage your family" thing - V. Unpleasent indeed. The thought that there may be a tracker in his cars that would allow anyone (Gov. included) to be able to track the whereabouts of his vehicles, which may contain his wife and kids, is something he isn't very keen on now.
Sure, it's a Gov system and we all know how secure they are, but even without the added pressure of the blackmailers, what if someone fancies burgling your house and can know when all your cars are out and are moving away from home. They could get a nice SMS message to inform them when the first of your vehicles starts heading home so they know when to vacate with the loot!
Perhaps a bit on the "tinfoil hat" side of things, but we've had people working at the DVLC selling info to people and the rewards for this info could be even higher!
Eclectic beats from Leeds, UK
handmadehands.co.uk
I agree. How did Robin go from zipcar to mesh?
I had an MBA friend who wanted to market a cordless hairdryer. I just shook my head.
Robin, I have a good cause, help me out. Its called wikispeeedia . It helps people be safe if they want to.
We need to take back OUR roads.
Don't want my car part of anything not under my complete control and with no monitoring by anyone other then myself by sitting in the drivers seat.
---- Booth was a patriot ----
Its nothing new, the idea of cars talking to each other to transmit road conditions, keep a certain distance, allow faster fluid road usage, impose road-travel pricing, etc have been around for a while.
http://www.today.ucla.edu/portal/ut/070508_network-on-wheels.aspx
http://reviews.cnet.com/4520-10895_7-6733591-1.html
There are probably other reasons to have it, from the 2nd link:
Google is also taking a strong interest in this technology. Why would an Internet search company be interested in car technology? Because it wants to extend its reach into your car. And where Google goes, Yahoo and Microsoft are likely to follow. Right now, navigation systems have static databases of restaurants, gas stations, and other businesses. A vehicle communication infrastructure could make that dynamic by sending requests for local restaurants, for example, over the network, with results coming back from Google, Yahoo, or any other online database.
so - safety, taxation, and advertising. I suppose it would also make stealing your car nigh on impossible, and it might help with congestion too.
OBD2 compliant automobiles sold in the USA are required by the DOT to have black boxes (for lack of a better term) that record vehicle speed, brake status, RPM's, and the other assortment of telemetry available in a modern engine for the sole purpose of the police using it against you in post-crash investigations. This is a proven fact.
No, it is not. The box stores "telemetry" data (i.e. sensor state) when ANY fault occurs. Also, if you're driving within the legal requirements, then the data can be used by your defense on your behalf during post-crash investigations.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
We must be as stealthy as rats in the wainscoting of their society. It was easier in the old days, of course, and society had more rats when the rules were looser, just as old wooden buildings have more rats than concrete buildings. But there are rats in the building now as well. Now that society is all ferrocrete and stainless steel there are fewer gaps in the joints. It takes a very smart rat indeed to find these openings. Only a stainless steel rat can be at home in this environment ; ).
"Kill 'em all and let Root sort 'em out"
Anyway, my city uses natural gas or something to fuel the buses.
Maybe bio-gas? In Stockholm, most inner-city buses use ethanol as fuel, except for some that use bio-gas. The bio-gas comes from the primary sewage treatment plant in the city. Diesel is only used for buses that have a significant part of their route outside the inner-city limits, with the stated goal that all buses should be using renewable fuels within just a few years.
Except that housing in suburbs is spread out over large areas -- very different than housing in urban areas. You CAN include suburbs in mass transit solutions, it's just that to do it in a way people will actually use will cost a crap load more money because you'd have to have many more stations.
"The tree of liberty must be refreshed from time to time with the blood of patriots and tyrants." ~Thomas Jefferson
And then, as if to illustrate my point, this shows up on Slashdot's front page today:
http://tech.slashdot.org/article.pl?sid=09/05/10/1549252&art_pos=8
Yeah. Paranoid. Right.
Mass transit busses remain on the roads all day and most of the night, unlike a commuting auto, which is normally used an hour or two per day
What a ridiculous comparison. If this is your logic, then you really having nothing to add to the subject at hand. We are not comparing 1 car vs 1 bus. We are comparing ALL cars against ALL buses, since it is moronic to look at it otherwise.