You realize that all states subsidize businesses and utilities already, right? Low interest loans, access to the bond market, tax deductions, heck, some companies get to keep all the sales tax they collect.
These kinds of things make sense at a scale that most people can't (or won't) think about. You can get an infrastructure built for less if you are willing to commit to funding a larger system over building multiple small systems. Suppliers will lower prices on bulk orders or provide long-term price guarantees, much fewer lawyer fees from repeated negotiations, less time lost in negotiations which allows inflationary forces to increase costs, etc, etc.
So yes, the cost of this installation is subsidized by everyone else in NJ but everyone else will ultimately benefit by reductions (or lack of increases) in their power bills when infrastructure improvements are deferred or canceled entirely due to reduced load on the power system due to those subsidized installations.
This program is actually a "triple threat" scenario. It 1) stimulates the economy since in general every $1 spent on a project actually gets spent multiple times. 2) It is a Capital improvements that lower costs and 3) it benefits the overall environment by lowering hydrocarbon emissions from coal plants.
"Real" cloud computing is supposed to be based on a mesh of geographically diverse, redundant servers each carrying various subsets of the data. Think RAID5 for servers, with each partition located in a different part of the world and on different networks.
Which means it is nothing more than an internet based service with five 9s of reliability and availability.
However it is an *expensive* internet based service so it needs a new moniker. But without a "Cloud Computing Consortium" with ownership of the trademark "cloud computing" to enforce correct usage, there's nothing to prevent everyone and their dog from using the term incorrectly for any "always connected" application.
The problem with all this is that its almost impossible for an end user to know for sure if someone really has a proper cloud application until something fails. If there is a total failure of a site and no one notices, you've got a working cloud. If people lose data or functionality, you don't.
That's the real killer. Even if you had all your data loaded on the phone, lose power and poof! With no mechanism to make local backups, you're utterly at the mercy of the cloud.
I've got a Pre, which is a cloud device, but if my battery dies the same time as the remote servers my data's safe for quite a while. Once the battery recharges I can get one of the sync apps to offload my data. If I were more paranoid I'd get one now but I try to make my own archives straight from Google, webmail, et al.
Look at the last paragraph in my post. I don't think in 2007 Linux had any wins in the trading market. While MS didn't either, I bet you that MS was willing to spend a big chunk of change to build a demo.NET exchange and run recorded LSE transactions through it to prove the speed boost.
There's a big difference between a known speed improvement and a potential one. If the.NET demo system was able to handle the worst day the LSE's older system had handled, and done so much faster, then it was a reasonable upgrade.
If the Linux exchanges in place have handled as much traffic as the day the.NET exchange failed while doing it faster, Linux is a reasonable upgrade.
You may think that the 6.75x speed increase of linux over.NET is a lot but keep in mind that.NET provided a 51.8x performance improvement (140ms/2.7ms) in 2007 when it was rolled out.
The article about the 2008 LSE crash says that before the LSe migrated to.net in June of 2007, trades took 140ms..NET reliability turned out to be less than ideal but at the time of the migration the speed boost was little short of phenomenal.
I don't think in 2007 Linux had any wins in the trading market. Which is an advantage the linux platform they bought has now; a proven track record running stock exchanges.
Dion Almaer and Ben Gailbraith (worked on Bespin at Mozilla) are now Palm employees and in charge of developer relations. Obviously, Palm is taking the problems with their app store and developer program seriously.
And you know that it really breaks down to a distribution arguement, right? JWZ can distribute his apps through other channels all he wants. But for the official store, which is still in beta and therefore not a done deal, you have to agree to their hoops. Their hoops can, and have, changed.
The Pre's easy to unlock (enter the konami code in the global search) and then get the installer utility. Pretty much the same as Android.
Is the Pre as unlocked as Centros and Treos? No, it's not. And if you ask me, it's because 10-15% of PalmOS software was buggy and caused problems but end users didn't say "hey, Widgets2005 makes my phone unstable" they said "This phone suxxorz and crashes all the time. Stupid Palm." My boss had the same Treo 650 I did but his crashed left and right thanks to some crapware he installed. I had occassional crashes on my 650 but I knew how to check the logs to see which program caused it and I'd kill that program.
Palm saw that most phones have virtually no ability to install apps and the ones that did (iPhone, Android) are tied to an app store. Most people are happy with centralized distribution systems and Palm made sure the walls of their garden were really low but rather wide, so that anyone can get out but that they can't be unaware that they are going "off reservation" with all the insecurity that entails.
This is for situations where you need lots of processes running but that those processes are either easily completed, are low-impact, or limited by bandwidth or the user. Web servers love lots of cores
On gaming you could separate the game into a user environment thread, a physics thread, an object management thread, a pair of AI threads, and still have a core left over for general OS activity.
I know that in theory compilers could also pull loops and modules out to separate threads but I haven't the foggiest clue if that really happens.
MS gets to dictate what drivers it will include or support, so Windows tends to run on specific ranges of equipment. WIndows gets to have a splash screen that says "this equipment is not supported by this version of Windows. Please contact Microsoft to purchase an appropriate version of Windows." Linux, meanwhile, gets dropped on anything and everything.
So Linux can run just about anywhere. Whether it does so efficiently or not depends on what hardware you have.
I'm a geeky person but I don't run Linux as my primary OS. There are too many times I need a Windows/Mac-only program, either for my job or for some 3rd party piece of technology (cellphone, PDA, DVR, etc).
However I'd recommend getting a LiveLinux DVD or USB stick for any laptop. By definition laptops get moved around and are more likely to experience disk drive failures. Inevitably, those failures often happen while you are on the road, with nothing other than your laptop bag. With a LiveLinux you have a way to pull data off a corrupted windows install and to remain functional with a jacked up hard drive.
Typing is muscle memory just like playing music. Guitarists don't consciously think "E-G-A-E-G-B/A-E-G-E-E-G" any more than a typist thinks "s-m-o-k-e- -o-n- -t-h-e- -w-a-t-e-r". (Well, bassists might but we're known to be pretty dense.)
I actually had trouble typing the hyphenated parts as my hands initially would spell out the whole words.
Have you never had a case where your fingers know your password but you don't? Happens to me all the time.
He needs a pump in the lab. If you used a river delta, where there's a natural water flow, you only need a series of diversion gates; one to let fresh water in, one to let seawater in.
He states that in theory you could capture 1.6KJ/Litre of water. 1 Liter water =~ 1 kg. Assuming you want 1m of elevation difference (it simplifies the math and that much head is a pretty solid flush), that's 9.81 J/liter at 100% pump efficiency. Assuming ~30% total pump efficiency, you're at ~30 J/liter. Now say that he has to pump both fresh and salt water, so it takes a total of 60 J of pumping per liter of fresh water reaction. If his system can reach 50% theoretical efficiency (0.8 KJ/liter) then you'd generate ~0.74 KJ/L of fresh water.
Using my example above of diverting ~10% of the mississippi river, Mississippi = 572,000 ft^3/s * 28 L/ft^3 x 10% x 0.74 KJ =~ 1.2GJ/s =~ 1.2GW.
I'd say a 1.2GW power plant is a pretty nifty goal.
Of course at the moment he's generating 0.00005J/liter in his proof-of-concept unit, so that 50% theoretical may be lofty, but with ~60 J/L of overhead, he starts positive power production at 4% theoretical. Not sure what percentage he needs to produce more power than manufacturing the the carbon and other plant facilities requires.
You really want to confuse them, change your phrase to "clubbed *WITH* a wet baby seal!" The double take is well worth it.
I wish I had the Flash skill to make an animation of an eskimo weilding a pair of baby seals like nunchucks. Possibly to club chickens to death for Colonel Sanders.
The article actually has an interesting addendum at the end that explains it, albeit in an interesting vernacular.
In short, salt water is ionic. A small initial electric charge is given to the two pieces of carbon (one positive, the other negative). The sodium and chlorine ions migrate to the respective carbon and thanks to the very high surface area of activated carbon, you get a very high quantity of ions. The water source then switches to fresh water. Elecrostatic force tries to keep the sodium & chlorine ions near the carbon but diffusion pulls them away. The work done to pull the ions away is what generates the power.
The inventor that it can generate as much as 1.6KJ / Liter of fresh water. If we diverted 10% of the Missisippi River's outflow into one of these facilities you get ~2.6GW of more or less continuous power. (Mississippi = 572,000 ft^3/s * 28.32 L/ft^3 x 10% x 1.6KJ = 2.6GJ/s = 2.6GW)
Well, if you use a 4 sq. foot lens over a 1 sq. inch cell you have ~576x brightness, assuming a perfect lens and no atmospheric absorption. More realistically you'd get closer to 500x brightness.
So if you use cheap, non-imaging fresnel lenses as concentrators you can get by with far fewer cells. At 28.3% efficiency, that 1 sq.in. cell will put out up to 0.2Kw, depending on location.
You can argue that they are converters rather than generators but that's just being pedantic.
However, one thing Palm did not duplicate is the Powervr chip. Apple was reported to be licensing this chip in 2008.
Palm didn't duplicate the VR, that's what Texas Instruments is for. Palm Pre uses the TI OMAP processor, which is the ARM Cortex A8 + PowerVR SGX integrated together. See that "Power VR" part? Yeah, the Palm Pre's out today and has essentially the same chip that the 3GS will use next week.
Which means at the moment, AFAIK, the Pre has the most powerful mobile phone CPU on the market. The Pre leapfrogged the existing iPhone models and the 3GS will be matching the Pre. Who was first? Meh, not worth arguing since I don't really think a week makes a big difference either way but the Pre is not in any way at a technical disadvantage to the 3GS.
Go read a few reviews. The Pre is generally seen to perform comparably to the 3GS with similar battery life. Not identical, given the differences in apps and usage patterns, but close enough that neither is clearly ahead of the other.
Well, until someone starts selling extended Pre battery packs.
Engineers can they just generally don't. I spent three years working at a library so I have a fondness for good organization systems.
I was the file system nazi at my last company, a civil engineering firm. I was hired for engineering and IT support right as they started implementing a standard. The nazi-ism started by marking every directory in the existing file store read-only on every project that was complete, according to the accountants.
Then "create directory" permissions were limited to senior project managers and their one administrative assistant. I set up a script to check for new directories every day and I'd email anyone who didn't follow protocol. I pre-seeded the directory structure by getting the list of open project numbers from finance so in theory, everything billable already had a home waiting for it. For new projects I simplified things by creating little widget that asked for a project number and the contract name and it created the directory tree.
We created a separate volume that contained data that was not project specific but may be needed across multiple projects. I.e. the various CAD standards (national, Corps of Engineers, DoD, DoT, etc) along with company/client logos, all the stock patterns/icons for the various CAD programs, etc. All the CAD programs were set to point to that shared directory by default to encourage the worker-bees to put shared data there so they wouldn't have to set project-specific directory over-rides.
That directory allowed everyone to add data but only the CAD/marketing/PR/QA managers could delete/overwrite files. A report was generated monthly and send to the managers that listed files with similar names and extensions to make sure we didn't wind up with 25 versions of one logo or hatch pattern.
This was staff-intensive but that's because capital expenditures were the devil since they couldn't be charged easily to a project. File management, however, is something that was billable.
Multitasking is a big, big benefit, assuming you're an internet-over-phone user. (And really, that's the target market. Not iphone users, net-phone users.) Right now if I surf the net on my phone it takes 5-10 seconds after I hit a link for a simple page to render and up to 30 seconds for an image laden one. I'm twiddling my thumbs (literally) in the meantime. With the WebOs I can fire up three or four web instances to cycle between loaded content, open up a game or use another app and get the web pages later.
In a concrete example, I can have the turn-by-turn voice navigation system giving me directions while I GoogleMap for restaurants, pull down restaurant reviews, and send messages to friends saying where to meet. (The GSM one coming out later this year will let you use internet and voice at the same time, the CDMA one can't, but that's a technology limiter)
The WebOs will have more open development than the iPhone simply because there is an O'Reilly book on programming for the WebOS that is already in production (http://my.safaribooksonline.com/9780596802097), the SDK has been rolled out to a small group of 3rd party developers with more on the way (http://developer.palm.com), and 3rd party apps are around at launch. Compare that to the release of the iPhone SDK (SDK? What SDK? We don't need no stinking SDK. Local apps are redundant. Oh, wait, here's an SDK. And an App store.)
Many of the others that are "better" is a matter of opinion that will have to be determined by person by person (aka the market). Basing it on bullet points and other reviewers, most of the stock apps are about on par with the iPhones. Some are a bit better, some a bit worse. Some, like the PIM, are not quite as good as the older PalmOS in some ways but have other features that trump everything else. E.g. support for synching multiple calendars with color coding identifiers is wonderful, the accordion-fold time compression is another good idea, but not having ways to categorize contacts or memos is a step backwards.
Much of "better" will come down to how well the concepts mesh with the way people work and/or think that they'll use the device.
Note that I am totally jonesing for a Pre but I'm waiting 'til a) my Sprint contract is up in August for the extra discounts and b) Docs2Go is released b/c I have something like 200,000 words worth of text documents I authored on my Centro and the existing memos app won't meet my need.
... when you are talking about a brand new type of device - not a smart phone but a computer that functions as a smart phone and media player -
There is nothing the iPhone does that any of the many preceding PalmOs cellphones don't support, excluding UI related items like multitouch or tying into other Apple products. They do email, web, video, audio, document editing, GPS, and tons of 3rd party apps.
I've never had a symbian phone so I can't say for certain that they are as fully functioned as the PalmOS devices.
The Treos evolved out of the Palm Pilot, which was itself a better PDA than the Newton. (yes, the newton was more powerful but it was a brick, the Pilot got the job done in a friendly form factor)
The iPhone was the *best* incarnation of a consumer smartphone after it got a few firmware updates and could reliably make phone calls but it was far from a "a new kind of device".
I doubt the Pre will topple the iPhone's ivory tower b/c the iPhone UI has it's fans and rightfully so. However as a person who only reluctantly started using the thumb-boards, I must say that I now can't conceive of not having a keyboard on my phone now. And for that market, I suspect the Pre will complete in a "separate but equal" fashion.
The HSR plan wouldn't expect you to. I quote the executive summary:
President Obama proposes to help address the Nation's transportation challenges by investing in an efficient, high-speed passenger rail network of 100- to 600-mile intercity corridors that connect communities across America.
For your 1500 mile trip to Dallas you should take a plane. But would you take a plane to Phiadelphia? What about DC? That's where the HSR is intended to be used.
The federal government has invested more than $10B/year in air travel since 1990. Let me repeat that. Each year since 1990, the federal government has invested at least 10 Billion U.S. dollars in air travel. For 2007 it was about $13B based on the graph in the HSR plan on page 13. (http://www.fra.dot.gov/Downloads/RRdev/hsrstrategicplan.pdf)
I'm going to use passenger ton-miles to distribute that cost so that I'm using real numbers. If we spread $13B over the 84B passenger ton-miles from 2007 listed on that website it that comes out to $0.15/ton-mile.
The website said there were 769M passenger-flights, so the average flight over 84B passenger ton-miles was was ~108 ton-miles. So the federal government actually subsidized the average one-way flight by $16.70.
That means the federal government paid for 16.7% of your hypothetical $100 average flight. Ramp that percentage up or down based on actual average flight prices.
If you read the first PDF, hsrstrategicplan.pdf, go to page 18, under Section 301, 501, 502 you can see that up to 80% of funding may come from the government. The exact amount depends on how closely it matches the goals of the HSR plans and/or benefits other types of rail service.
While nothing excludes provide companies from getting involved, they MUST involve the states and have the project added to that state's Rail Plan. This means the project will have a stronger local component and firmer commitments by everyone involved.
Applications are due August 2009 with a draft national rail plan out in October 09. They plan on at least two phases of projects, with the 2nd phase accepting new project applications starting January 2010.
I live outside of Louisville and would love to be able to get to Indy in an hour or Chicago in 3 hours or less. I would be much more likely to go to out-of-town concerts and events if I didn't have to spend hours behind the wheel. Being able to nap in a train and especially being able to stretch my legs a bit without stopping the car would be idea.
Driving ~6hours to Chicago is not appealing and after getting to the airport early, the 1.5 hour flight turns into the same 3 hours as a train ride.
The other thing is that trains are rarely grounded by fog or storms. I can't count the number of my flights that were delayed by weather.
Also, while a multi-chip design reduces the probability of a device failing outright, it dramatically increases the probability of a failure.
I didn't make it clear that I was referring to the manufacturing side of things. I meant multiple chips reduced the chance of failures during manufacturing making the whole product unsellable. The more transistors to the package, the greater chance that some of them will be bad off the line. If the package can't tolerate any transistor failures and the cost per failed unit is high enough, you're better off building component chips indvidually then joining them on a PCB after validation.
Plus, many flashdrive manufacturers are assemblers and not chip fabs. They buy flash and controllers from various fabs and install them on PCBs in the apropriate combination to get their sizes. An external controller makes it easier to switch between producing 8GB, 16GB, or 64GB flashdrives since they only have to change the size and/or quantity of the flash chips.
Given the volatility in the flash market both to size and cost, I'm not sure it is financially viable to produce many memory+integrated controllers for anything but the largest bulk orders.
I haven't bought a retractable flash drive. Does the whole PCB slide within the housing or is there a flexible ribbon connecting the PCB to the USB connector? The former seems like it would be cheaper and more durable but the latter seems lazier and lazy seems to rule the day.
I doubt that CFLs put out fewer lumens/watt than incandescents, although they probably do lose some brightness over time. I've seen cheap CFLs die early deaths or fail to start on a prompt manner or require a couple minutes to warm up but I've never seen a 100W equivalent CFL put out less light than a 25W incadescent.
I can't even conceive of a mechanism that would allow that. If the gas is flourescing, it's pretty much putting out the appropriate amount of lumens, albeit in UV. While I've seen larger tubes "strobe", the CFLs seem to be immune to that, plus the flicker would drive you mad. Alternately the actual flourescent coating that converts the UV discharge to white light could have been worn away but that's something I've never heard of happening as that's a mature manufacturing process.
If a cell fails, you can't read or write that cell.
This is a silent failure, much like hard drives marking blocks as bad. Capacity is reduced without any obvious signs. Not sure if OS tools can recognize it unless the controller reports bad cells as bad blocks. This will eventually result in "disk full" messages when there appears to be space on the drive. Reformatting won't recover the cells but it will likely result in your OS being aware of the flash's reduced capacity.
If a gate fails in a page, you lose access to the page.
Very similar to above, but larger amounts of data. I want to say there's 64 cells to the page but don't take that as gospel.
If a gate fails in the overall control logic, you lose access to the whole device.
Hello failed/unreadable/size 0 disk error. The data storage mechanism is intact but there's no way to access them.
As people stated above, a lot of the time it is not the failure of a transistor so much as a trace or solder point failing. If you know your device has been abused physically, you can try the low-tech approach of gently squeezing or bending the stick while it's in the USB port (use an extension cable so you don't damage your mobo!!) to try and get the contacts to reconnect long enough to retrieve data. If that fails you can pop the case apart and use a magnifying glass to look for breaks in the solder or traces; if you're handy with a soldering iron you can try to bridge the connection. Again, temporary fix.
Is there something I'm missing? Did you think there were oil changes or brake shoes? It's one silicon chip with metal on it.
Actually most of them are several silicon chips; one controller plus a variable amount of memory chips. The increase in traces and board assembly is offset by the ability to reuse components and the overall design while memory chip prices fall. It also cuts down on the impact of failed chips, since you aren't losing controller+memory for one bad gate on the controller.
True but the incandescent has a 107VA grid load (100W actual + 7VA apparent) while the CFL will have 39VA grid load (25W actual + 14VA apparent). That's still a ~60% reduction in load.
My co-op is giving out CFLs to reduce the load of existing users allowing for new growth without more generation. Distribution upgrades may be necessary to counter the influence of the power factor but they may be dealt with as part of routine component replacements and the cost can be carried over to the new construction. The big thing is that if ~60% of power generation is used for lighting, CFLs could free up ~30% of their total generation capacity. That number will never happen until street lights and other area lights are switched to something higher efficiency but getting the equivalent of 10% generation back is a big deal.
Slashdot Mirror
You realize that all states subsidize businesses and utilities already, right? Low interest loans, access to the bond market, tax deductions, heck, some companies get to keep all the sales tax they collect.
These kinds of things make sense at a scale that most people can't (or won't) think about. You can get an infrastructure built for less if you are willing to commit to funding a larger system over building multiple small systems. Suppliers will lower prices on bulk orders or provide long-term price guarantees, much fewer lawyer fees from repeated negotiations, less time lost in negotiations which allows inflationary forces to increase costs, etc, etc.
So yes, the cost of this installation is subsidized by everyone else in NJ but everyone else will ultimately benefit by reductions (or lack of increases) in their power bills when infrastructure improvements are deferred or canceled entirely due to reduced load on the power system due to those subsidized installations.
This program is actually a "triple threat" scenario. It 1) stimulates the economy since in general every $1 spent on a project actually gets spent multiple times. 2) It is a Capital improvements that lower costs and 3) it benefits the overall environment by lowering hydrocarbon emissions from coal plants.
"Real" cloud computing is supposed to be based on a mesh of geographically diverse, redundant servers each carrying various subsets of the data. Think RAID5 for servers, with each partition located in a different part of the world and on different networks.
Which means it is nothing more than an internet based service with five 9s of reliability and availability.
However it is an *expensive* internet based service so it needs a new moniker. But without a "Cloud Computing Consortium" with ownership of the trademark "cloud computing" to enforce correct usage, there's nothing to prevent everyone and their dog from using the term incorrectly for any "always connected" application.
The problem with all this is that its almost impossible for an end user to know for sure if someone really has a proper cloud application until something fails. If there is a total failure of a site and no one notices, you've got a working cloud. If people lose data or functionality, you don't.
That's the real killer. Even if you had all your data loaded on the phone, lose power and poof! With no mechanism to make local backups, you're utterly at the mercy of the cloud.
I've got a Pre, which is a cloud device, but if my battery dies the same time as the remote servers my data's safe for quite a while. Once the battery recharges I can get one of the sync apps to offload my data. If I were more paranoid I'd get one now but I try to make my own archives straight from Google, webmail, et al.
Umm, no, it doesn't.
Look at the last paragraph in my post. I don't think in 2007 Linux had any wins in the trading market. While MS didn't either, I bet you that MS was willing to spend a big chunk of change to build a demo .NET exchange and run recorded LSE transactions through it to prove the speed boost.
There's a big difference between a known speed improvement and a potential one. If the .NET demo system was able to handle the worst day the LSE's older system had handled, and done so much faster, then it was a reasonable upgrade.
If the Linux exchanges in place have handled as much traffic as the day the .NET exchange failed while doing it faster, Linux is a reasonable upgrade.
You may think that the 6.75x speed increase of linux over .NET is a lot but keep in mind that .NET provided a 51.8x performance improvement (140ms/2.7ms) in 2007 when it was rolled out.
The article about the 2008 LSE crash says that before the LSe migrated to .net in June of 2007, trades took 140ms. .NET reliability turned out to be less than ideal but at the time of the migration the speed boost was little short of phenomenal.
I don't think in 2007 Linux had any wins in the trading market. Which is an advantage the linux platform they bought has now; a proven track record running stock exchanges.
Dion Almaer and Ben Gailbraith (worked on Bespin at Mozilla) are now Palm employees and in charge of developer relations. Obviously, Palm is taking the problems with their app store and developer program seriously.
And you know that it really breaks down to a distribution arguement, right? JWZ can distribute his apps through other channels all he wants. But for the official store, which is still in beta and therefore not a done deal, you have to agree to their hoops. Their hoops can, and have, changed.
The Pre's easy to unlock (enter the konami code in the global search) and then get the installer utility. Pretty much the same as Android.
Is the Pre as unlocked as Centros and Treos? No, it's not. And if you ask me, it's because 10-15% of PalmOS software was buggy and caused problems but end users didn't say "hey, Widgets2005 makes my phone unstable" they said "This phone suxxorz and crashes all the time. Stupid Palm." My boss had the same Treo 650 I did but his crashed left and right thanks to some crapware he installed. I had occassional crashes on my 650 but I knew how to check the logs to see which program caused it and I'd kill that program.
Palm saw that most phones have virtually no ability to install apps and the ones that did (iPhone, Android) are tied to an app store. Most people are happy with centralized distribution systems and Palm made sure the walls of their garden were really low but rather wide, so that anyone can get out but that they can't be unaware that they are going "off reservation" with all the insecurity that entails.
This is for situations where you need lots of processes running but that those processes are either easily completed, are low-impact, or limited by bandwidth or the user. Web servers love lots of cores
On gaming you could separate the game into a user environment thread, a physics thread, an object management thread, a pair of AI threads, and still have a core left over for general OS activity.
I know that in theory compilers could also pull loops and modules out to separate threads but I haven't the foggiest clue if that really happens.
MS gets to dictate what drivers it will include or support, so Windows tends to run on specific ranges of equipment. WIndows gets to have a splash screen that says "this equipment is not supported by this version of Windows. Please contact Microsoft to purchase an appropriate version of Windows." Linux, meanwhile, gets dropped on anything and everything.
So Linux can run just about anywhere. Whether it does so efficiently or not depends on what hardware you have.
I'm a geeky person but I don't run Linux as my primary OS. There are too many times I need a Windows/Mac-only program, either for my job or for some 3rd party piece of technology (cellphone, PDA, DVR, etc).
However I'd recommend getting a LiveLinux DVD or USB stick for any laptop. By definition laptops get moved around and are more likely to experience disk drive failures. Inevitably, those failures often happen while you are on the road, with nothing other than your laptop bag. With a LiveLinux you have a way to pull data off a corrupted windows install and to remain functional with a jacked up hard drive.
Typing is muscle memory just like playing music. Guitarists don't consciously think "E-G-A-E-G-B/A-E-G-E-E-G" any more than a typist thinks "s-m-o-k-e- -o-n- -t-h-e- -w-a-t-e-r". (Well, bassists might but we're known to be pretty dense.)
I actually had trouble typing the hyphenated parts as my hands initially would spell out the whole words.
Have you never had a case where your fingers know your password but you don't? Happens to me all the time.
He needs a pump in the lab. If you used a river delta, where there's a natural water flow, you only need a series of diversion gates; one to let fresh water in, one to let seawater in.
He states that in theory you could capture 1.6KJ/Litre of water. 1 Liter water =~ 1 kg. Assuming you want 1m of elevation difference (it simplifies the math and that much head is a pretty solid flush), that's 9.81 J/liter at 100% pump efficiency. Assuming ~30% total pump efficiency, you're at ~30 J/liter. Now say that he has to pump both fresh and salt water, so it takes a total of 60 J of pumping per liter of fresh water reaction. If his system can reach 50% theoretical efficiency (0.8 KJ/liter) then you'd generate ~0.74 KJ/L of fresh water.
Using my example above of diverting ~10% of the mississippi river, Mississippi = 572,000 ft^3/s * 28 L/ft^3 x 10% x 0.74 KJ =~ 1.2GJ/s =~ 1.2GW.
I'd say a 1.2GW power plant is a pretty nifty goal.
Of course at the moment he's generating 0.00005J/liter in his proof-of-concept unit, so that 50% theoretical may be lofty, but with ~60 J/L of overhead, he starts positive power production at 4% theoretical. Not sure what percentage he needs to produce more power than manufacturing the the carbon and other plant facilities requires.
You really want to confuse them, change your phrase to "clubbed *WITH* a wet baby seal!" The double take is well worth it.
I wish I had the Flash skill to make an animation of an eskimo weilding a pair of baby seals like nunchucks. Possibly to club chickens to death for Colonel Sanders.
The article actually has an interesting addendum at the end that explains it, albeit in an interesting vernacular.
In short, salt water is ionic. A small initial electric charge is given to the two pieces of carbon (one positive, the other negative). The sodium and chlorine ions migrate to the respective carbon and thanks to the very high surface area of activated carbon, you get a very high quantity of ions. The water source then switches to fresh water. Elecrostatic force tries to keep the sodium & chlorine ions near the carbon but diffusion pulls them away. The work done to pull the ions away is what generates the power.
The inventor that it can generate as much as 1.6KJ / Liter of fresh water. If we diverted 10% of the Missisippi River's outflow into one of these facilities you get ~2.6GW of more or less continuous power. (Mississippi = 572,000 ft^3/s * 28.32 L/ft^3 x 10% x 1.6KJ = 2.6GJ/s = 2.6GW)
Well, if you use a 4 sq. foot lens over a 1 sq. inch cell you have ~576x brightness, assuming a perfect lens and no atmospheric absorption. More realistically you'd get closer to 500x brightness.
So if you use cheap, non-imaging fresnel lenses as concentrators you can get by with far fewer cells. At 28.3% efficiency, that 1 sq.in. cell will put out up to 0.2Kw, depending on location.
You can argue that they are converters rather than generators but that's just being pedantic.
However, one thing Palm did not duplicate is the Powervr chip. Apple was reported to be licensing this chip in 2008.
Palm didn't duplicate the VR, that's what Texas Instruments is for. Palm Pre uses the TI OMAP processor, which is the ARM Cortex A8 + PowerVR SGX integrated together. See that "Power VR" part? Yeah, the Palm Pre's out today and has essentially the same chip that the 3GS will use next week.
Which means at the moment, AFAIK, the Pre has the most powerful mobile phone CPU on the market. The Pre leapfrogged the existing iPhone models and the 3GS will be matching the Pre. Who was first? Meh, not worth arguing since I don't really think a week makes a big difference either way but the Pre is not in any way at a technical disadvantage to the 3GS.
Go read a few reviews. The Pre is generally seen to perform comparably to the 3GS with similar battery life. Not identical, given the differences in apps and usage patterns, but close enough that neither is clearly ahead of the other.
Well, until someone starts selling extended Pre battery packs.
Engineers can they just generally don't. I spent three years working at a library so I have a fondness for good organization systems.
I was the file system nazi at my last company, a civil engineering firm. I was hired for engineering and IT support right as they started implementing a standard. The nazi-ism started by marking every directory in the existing file store read-only on every project that was complete, according to the accountants.
Then "create directory" permissions were limited to senior project managers and their one administrative assistant. I set up a script to check for new directories every day and I'd email anyone who didn't follow protocol. I pre-seeded the directory structure by getting the list of open project numbers from finance so in theory, everything billable already had a home waiting for it. For new projects I simplified things by creating little widget that asked for a project number and the contract name and it created the directory tree.
We created a separate volume that contained data that was not project specific but may be needed across multiple projects. I.e. the various CAD standards (national, Corps of Engineers, DoD, DoT, etc) along with company/client logos, all the stock patterns/icons for the various CAD programs, etc. All the CAD programs were set to point to that shared directory by default to encourage the worker-bees to put shared data there so they wouldn't have to set project-specific directory over-rides.
That directory allowed everyone to add data but only the CAD/marketing/PR/QA managers could delete/overwrite files. A report was generated monthly and send to the managers that listed files with similar names and extensions to make sure we didn't wind up with 25 versions of one logo or hatch pattern.
This was staff-intensive but that's because capital expenditures were the devil since they couldn't be charged easily to a project. File management, however, is something that was billable.
Multitasking is a big, big benefit, assuming you're an internet-over-phone user. (And really, that's the target market. Not iphone users, net-phone users.) Right now if I surf the net on my phone it takes 5-10 seconds after I hit a link for a simple page to render and up to 30 seconds for an image laden one. I'm twiddling my thumbs (literally) in the meantime. With the WebOs I can fire up three or four web instances to cycle between loaded content, open up a game or use another app and get the web pages later.
In a concrete example, I can have the turn-by-turn voice navigation system giving me directions while I GoogleMap for restaurants, pull down restaurant reviews, and send messages to friends saying where to meet. (The GSM one coming out later this year will let you use internet and voice at the same time, the CDMA one can't, but that's a technology limiter)
The WebOs will have more open development than the iPhone simply because there is an O'Reilly book on programming for the WebOS that is already in production (http://my.safaribooksonline.com/9780596802097), the SDK has been rolled out to a small group of 3rd party developers with more on the way (http://developer.palm.com), and 3rd party apps are around at launch. Compare that to the release of the iPhone SDK (SDK? What SDK? We don't need no stinking SDK. Local apps are redundant. Oh, wait, here's an SDK. And an App store.)
Many of the others that are "better" is a matter of opinion that will have to be determined by person by person (aka the market). Basing it on bullet points and other reviewers, most of the stock apps are about on par with the iPhones. Some are a bit better, some a bit worse. Some, like the PIM, are not quite as good as the older PalmOS in some ways but have other features that trump everything else. E.g. support for synching multiple calendars with color coding identifiers is wonderful, the accordion-fold time compression is another good idea, but not having ways to categorize contacts or memos is a step backwards.
Much of "better" will come down to how well the concepts mesh with the way people work and/or think that they'll use the device.
Note that I am totally jonesing for a Pre but I'm waiting 'til a) my Sprint contract is up in August for the extra discounts and b) Docs2Go is released b/c I have something like 200,000 words worth of text documents I authored on my Centro and the existing memos app won't meet my need.
... when you are talking about a brand new type of device - not a smart phone but a computer that functions as a smart phone and media player -
There is nothing the iPhone does that any of the many preceding PalmOs cellphones don't support, excluding UI related items like multitouch or tying into other Apple products. They do email, web, video, audio, document editing, GPS, and tons of 3rd party apps.
I've never had a symbian phone so I can't say for certain that they are as fully functioned as the PalmOS devices.
The Treos evolved out of the Palm Pilot, which was itself a better PDA than the Newton. (yes, the newton was more powerful but it was a brick, the Pilot got the job done in a friendly form factor)
The iPhone was the *best* incarnation of a consumer smartphone after it got a few firmware updates and could reliably make phone calls but it was far from a "a new kind of device".
I doubt the Pre will topple the iPhone's ivory tower b/c the iPhone UI has it's fans and rightfully so. However as a person who only reluctantly started using the thumb-boards, I must say that I now can't conceive of not having a keyboard on my phone now. And for that market, I suspect the Pre will complete in a "separate but equal" fashion.
Yeah, this is where I was going. I thought about saying "except for the whole fraud thing" but thought it was overly glib.
Mea culpa. Next time I shall be glib.
President Obama proposes to help address the Nation's transportation challenges by investing in an efficient, high-speed passenger rail network of 100- to 600-mile intercity corridors that connect communities across America.
For your 1500 mile trip to Dallas you should take a plane. But would you take a plane to Phiadelphia? What about DC? That's where the HSR is intended to be used.
The federal government has invested more than $10B/year in air travel since 1990. Let me repeat that. Each year since 1990, the federal government has invested at least 10 Billion U.S. dollars in air travel. For 2007 it was about $13B based on the graph in the HSR plan on page 13. (http://www.fra.dot.gov/Downloads/RRdev/hsrstrategicplan.pdf)
I'm going to use passenger ton-miles to distribute that cost so that I'm using real numbers. If we spread $13B over the 84B passenger ton-miles from 2007 listed on that website it that comes out to $0.15/ton-mile.
The website said there were 769M passenger-flights, so the average flight over 84B passenger ton-miles was was ~108 ton-miles. So the federal government actually subsidized the average one-way flight by $16.70.
That means the federal government paid for 16.7% of your hypothetical $100 average flight. Ramp that percentage up or down based on actual average flight prices.
Here's the actual plan documents:
http://www.fra.dot.gov/us/content/31
If you read the first PDF, hsrstrategicplan.pdf, go to page 18, under Section 301, 501, 502 you can see that up to 80% of funding may come from the government. The exact amount depends on how closely it matches the goals of the HSR plans and/or benefits other types of rail service.
While nothing excludes provide companies from getting involved, they MUST involve the states and have the project added to that state's Rail Plan. This means the project will have a stronger local component and firmer commitments by everyone involved.
Applications are due August 2009 with a draft national rail plan out in October 09. They plan on at least two phases of projects, with the 2nd phase accepting new project applications starting January 2010.
I live outside of Louisville and would love to be able to get to Indy in an hour or Chicago in 3 hours or less. I would be much more likely to go to out-of-town concerts and events if I didn't have to spend hours behind the wheel. Being able to nap in a train and especially being able to stretch my legs a bit without stopping the car would be idea.
Driving ~6hours to Chicago is not appealing and after getting to the airport early, the 1.5 hour flight turns into the same 3 hours as a train ride.
The other thing is that trains are rarely grounded by fog or storms. I can't count the number of my flights that were delayed by weather.
I didn't make it clear that I was referring to the manufacturing side of things. I meant multiple chips reduced the chance of failures during manufacturing making the whole product unsellable. The more transistors to the package, the greater chance that some of them will be bad off the line. If the package can't tolerate any transistor failures and the cost per failed unit is high enough, you're better off building component chips indvidually then joining them on a PCB after validation.
Plus, many flashdrive manufacturers are assemblers and not chip fabs. They buy flash and controllers from various fabs and install them on PCBs in the apropriate combination to get their sizes. An external controller makes it easier to switch between producing 8GB, 16GB, or 64GB flashdrives since they only have to change the size and/or quantity of the flash chips.
Given the volatility in the flash market both to size and cost, I'm not sure it is financially viable to produce many memory+integrated controllers for anything but the largest bulk orders.
I haven't bought a retractable flash drive. Does the whole PCB slide within the housing or is there a flexible ribbon connecting the PCB to the USB connector? The former seems like it would be cheaper and more durable but the latter seems lazier and lazy seems to rule the day.
I doubt that CFLs put out fewer lumens/watt than incandescents, although they probably do lose some brightness over time. I've seen cheap CFLs die early deaths or fail to start on a prompt manner or require a couple minutes to warm up but I've never seen a 100W equivalent CFL put out less light than a 25W incadescent.
I can't even conceive of a mechanism that would allow that. If the gas is flourescing, it's pretty much putting out the appropriate amount of lumens, albeit in UV. While I've seen larger tubes "strobe", the CFLs seem to be immune to that, plus the flicker would drive you mad. Alternately the actual flourescent coating that converts the UV discharge to white light could have been worn away but that's something I've never heard of happening as that's a mature manufacturing process.
This is a silent failure, much like hard drives marking blocks as bad. Capacity is reduced without any obvious signs. Not sure if OS tools can recognize it unless the controller reports bad cells as bad blocks. This will eventually result in "disk full" messages when there appears to be space on the drive. Reformatting won't recover the cells but it will likely result in your OS being aware of the flash's reduced capacity.
Very similar to above, but larger amounts of data. I want to say there's 64 cells to the page but don't take that as gospel.
Hello failed/unreadable/size 0 disk error. The data storage mechanism is intact but there's no way to access them. As people stated above, a lot of the time it is not the failure of a transistor so much as a trace or solder point failing. If you know your device has been abused physically, you can try the low-tech approach of gently squeezing or bending the stick while it's in the USB port (use an extension cable so you don't damage your mobo!!) to try and get the contacts to reconnect long enough to retrieve data. If that fails you can pop the case apart and use a magnifying glass to look for breaks in the solder or traces; if you're handy with a soldering iron you can try to bridge the connection. Again, temporary fix.
Actually most of them are several silicon chips; one controller plus a variable amount of memory chips. The increase in traces and board assembly is offset by the ability to reuse components and the overall design while memory chip prices fall. It also cuts down on the impact of failed chips, since you aren't losing controller+memory for one bad gate on the controller.
True but the incandescent has a 107VA grid load (100W actual + 7VA apparent) while the CFL will have 39VA grid load (25W actual + 14VA apparent). That's still a ~60% reduction in load.
My co-op is giving out CFLs to reduce the load of existing users allowing for new growth without more generation. Distribution upgrades may be necessary to counter the influence of the power factor but they may be dealt with as part of routine component replacements and the cost can be carried over to the new construction. The big thing is that if ~60% of power generation is used for lighting, CFLs could free up ~30% of their total generation capacity. That number will never happen until street lights and other area lights are switched to something higher efficiency but getting the equivalent of 10% generation back is a big deal.