I'm a community college physics teacher in the U.S., and I scrounge computer hardware to use in my lab classes. The school provides one Windows box per lab group, i.e., 7 computers for a class of 25 students. The trouble with that is, you get one student doing the graphs on the computer, and the rest of the group just sits there and watches. I've made a geekly hobby out of putting together decent Linux systems from garage sales, thrift shops, etc., to supplement what the school provides. It was interesting comparing the article with my own experiences back here in the developed world.
One thing to keep in mind is that the line between good and bad hardware is extremely fuzzy. I picked up an old 500 MHz e-Machines box recently at Good Will for $89, and with a $20 memory upgrade it makes a perfectly decent Linux machine, especially with a distro like xubuntu that's designed more for low-end hardware (xfce rather than gnome, abiword rather than OOo). Many people would have considered this machine too old to be useful, but it works fine for the application I need it for.
Similar deal with monitors. I actually find that cheap monitors are much, much harder to find than cheap computers. You don't see them much at thrift shops or swap meets, I guess because CRTs are heavy and bulky in relation to what you can sell them for. When I do get an old CRT, its mean time to failure is usually pretty darn short, probably 12-24 months. As far as I can tell, computer CRTs have a certain lifetime, and when you get your hands on a cheap one it's already near the end of that.
One thing that's absurd, when you view computers as potential solid waste, is the amount of air inside a tower case these days. On a low-end machine, the case can easily be 90% empty. It's the equivalent of going to McDonalds and having them serve you your little 99-cent hamburger in a styrofoam clamshell the size of a microwave oven. I'm hoping the Asus eee Box comes out soon, and Asus doesn't jack up the price. For $269, it could be a wonderful deal.
And by the way, if you're in Orange County, CA, and have a working monitor you're willing to donate, please email me at crowell08 at lightSPAMandISmatterEVIL dot com. I'll be more than happy to come and pick it up, and you can have the warm fuzzy feeling of knowing that you're helping me spread peace, love, and linux to my students.
My dispute with this line of reasoning is that we use an insignificant amount of oil for electricity generation purposes.
Where I live, in Southern California, I believe SCE generates our electricity from oil and natural gas. Oil comes from oilfields. Natural gas comes from oilfields and natural gas fields.
[...] coal and natural gas [...] receive some of the most marginal amounts of subsidy in the industry
The WSJ article you link to is talking about direct, overt government subsidies. Those aren't the subsidies I'm talking about.
What everyone seems to be waiting for is a cost-per-watt that is low enough so that ordinary people will decide to start buying them in large quantities without government subsidization. Suppose you're having a new house built: if you could install a ten or fifteen kilowatt solar plant and inverter for ten grand, you might figure it's worth it to borrow a little more money from the bank.
A couple of things to keep in mind here:
The cost per watt is already low enough that it makes sense for a lot of people, like me, to buy photovoltaics. It depends on what latitude you live at, how much sunny weather you get, which way your roof faces, how much shade there is on your roof, what the local price of electricity is, and what you expect the local price of electricity to be over the 25-year life of a photovoltaic system.
When you talk about government subsidies, you should do an apples-to-apples comparison with the alternative, which is typically electricity that comes from burning fossil fuels. Fossil fuels enjoy massive government subsidies here in the U.S. We've fought three extremely expensive wars recently in the middle east, and I don't think we would have been involved in any of those wars if there hadn't been oil there; my grandkids will be paying for my generation's deficit-funded oil wars. There's also a huge amount of environmental damage done by burning fossil fuels, and that damage affects both this generation and future generations. If people paid the real costs of that environmental damage up front, then gas would be a lot more expensive. In places like Europe that don't subsidize fossil fuels as much, gas costs about twice what it does in the U.S.
The big market for these are textbooks, that's why I didn't get one. If they offered digital versions of textbooks at a discount price this thing would kill.
Paper, printing, and binding are usually not a significant percentage of the cost of a textbook. Digital textbooks might have the advantage of cutting out the campus bookstore as a middleman, but I suspect that the publisher is not going to let the student have the money that would have gone to the store (typically about 27%). Another issue is returns, and the ability to buy and sell used copies, which probably wouldn't be possible.
It's fine that a lot of people seem to like the thing. Reasons I'm not interested:
$360 is way too much.
DRM.
The methods for importing PDF files sound like a hassle.
The TOU say you can't sell or give away your books.
There are only 145,000 books available. That sounds like a lot, but it's really not.
I can see how it could come in handy if you're on vacation and want to travel light, but IMO that's not nearly enough to overcome the negatives. I'll probably get an e-book reader in 2030 or something. There's no rush. First I want to see someone get it right.
Actually, although the Flash IDE is closed-source and proprietary, the SWF file format is now a published specification which others are free to implement.
I don't think the situation isn't quite that rosy. Although SWF is an open format, there are lots and lots of other bits and pieces that you need, such as the library of GUI widgets, and the various audio and video codecs. Those are all proprietary. (Some, e.g., mp3, are only proprietary because someone else owns a patent. However, Adobe hasn't bothered to implement support for free codecs such as vorbis and theora.) It's true that there are third-party, open-source compilers such as haxe. However, there's nothing that's compatible at the source-code level with Adobe's tools. Therefore if you're trying to work on an existing flash project, or trying to learn flash development from books, you really have no option other than Adobe's proprietary tools. The last time I checked, their tools also came with nasty EULAs, which, e.g., forbid reverse-engineering, creation of competing products, or use on a mobile device.
If you want encryption guaranteed against major governments you have to go with a one time pad.
Oh, please. You've done a good job of using impressive terms to sound like you know what you're talking about. If you want to talk about the real-world risks of having your crypto broken, then you need to consider all the real-world methods by which your crypto could be broken. It doesn't matter that a one-time pad can be theoretically proved to be invulnerable to certain kinds of attacks, to which various symmetric and asymmetric ciphers are theoretically vulnerable. What matters is the actual types of attacks that are practical and likely, and the actual problems you'll have in the practical implementation of a particular method. If you're using a one-time pad, then there are several obvious, well-known things that can go wrong: (1) you have to physically exchange the one-time pads, which may be difficult to do (and do securely) if the NSA is really following you everywhere, opening your mail, etc.; (2) both parties have to maintain the security of their own copies of the one-time pads, which may be difficult to do if the NSA is really determined to get them; (3) there is a tendency for users to get lazy and reuse a one-time pad, which then makes you vulnerable to certain kinds of attacks. Standard symmetric and asymmetric ciphers are more or less immune to these problems (#1: swapping passwords securely is a lot easier than swapping large amounts of binary data securely; #2: you can keep the password in your head instead of stuffing a keychain drive under your mattress; #3: no such issue). Yes, there are also certain kinds of attacks to which standard ciphers are vulnerable and one-time pads invulnerable (e.g., dictionary attacks on your password, shoulder-surfing,...) One-time pads are not magic pixie dust for cryptography. There is no magic pixie dust for cryptography. The good news is that we're living in a golden age of privacy, in the sense that you can legally, publicly get software to do encryption so good that essentially your main worry is no longer the encryption, it's the social/personal/legal issues surrounding its implementation.
Perhaps instead of malice or stupidity, it was simply "taking care of the biggest customer pool first."
Yeah. This also seems to be an example of a more general phenomenon with Linux support, which is that the same company will make completely contradictory statements about their own Linux support. In the earlier slashdot story, someone from Foxconn is directly quoted as saying 'it doesn't support Linux;' now they say they always intended to support Linux. The truth is probably that they never even thought about Linux support, and then when the issue was brought to their attention random representatives started saying random things off the cuff.
I've had a similar experience with Amazon's MP3 store. If you want to buy entire albums (as opposed to individual tracks), you have to use special downloading software that they supply. The software was initially only available in Windows and Mac versions, but pretty quickly they brought out Linux versions as well. Nowadays when you use your Linux box to shop for albumbs on their site, if you don't have the software installed your browser will detect that, and detect your OS as linux, and they'll generate a page for you offering links to download a linux version of the downloader. In fact, they even have it available in multiple versions for different linux distros. However, the linux downloader has been pretty buggy for me (and was also hard to get working properly on x64). I've had it working, then it broke, etc. I've done two calls to Amazon's tech support about this, and in both cases, the initial reaction was to tell me to do a bunch of stuff (with the usual confusion because the Indian tech support person gives Windows+IE instructions, and has never heard of Linux), and then when that didn't help they checked with someone else, who told them Linux wasn't supported. Never mind that they've had Linux versions of the software up on the site for months now.
I think part of the problem is that so many people in the hardware and software industries live in a 100%-Windows environment. It honestly never even occurs to them that anyone is running any other OS. (In the case of Foxconn, they're not making mac-compatible boards, so it's probably true that 99% of their boards are being used with Windows.) Then when the issue comes up, they just deal with it off the cuff. It's like asking them what their policy is on recycling cardboard -- they probably don't have one, and they don't see why it's important.
Another problem may be that in a Windows monoculture environment, many people don't understand what a standard really is. They think Windows and Word and IE are standards. Instead of developing for the relevant standard, some PHB makes the decision that they're going to target something proprietary, calling that a "standard," and they think of it as extra work to add support for anything else -- when in fact, it would have made more sense just to support the standard properly in the first place.
Yeah... have you ever tried working in one of the smaller cases?
I agree that for a machine I want to work on myself, a tower is great. Heck, even with a tower there are times when I'm scraping my knucles and cussing and swearing. But when's the last time you upgraded your transistor radio? We're headed into an era when PCs will sell for $50, and will be disposable.
But for other applications, "big" and "overpowered" is exactly what I want.
Sure, but that's only for those applications. You could compare it to a TV. A lot of affluent families these days own four TVs, but only one of those is the giant home-theater one. The rest are cheap consumer stuff from Costco, which they'll throw in a dumpster in 5 years.
The problem with this device is that it isn't that much cheaper than a full budget PC that will whack this into the ground.
What figure do you have in mind when you say "isn't that much cheaper?" It's basically impossible to walk into a retail store and buy a new PC for less than $400-500. The Everex gPC does have better specs, and is only $200, but it's only available online. For people who like the small form factor, the CherryPal might be preferable to the gPC. The gPC's Linux distro also sucks to high heaven, so for people who aren't technically adept enough to install a linux distro themselves, there really aren't any good options out there right now for less than $400-500.
I think we're currently in a period of marketplace chaos, and when the dust settles we'll find that a $1000 PC in a tower case seems about as archaic as a radio in a wooden case the size of a washing machine.
The biggest computer manufacturers are still selling machines in the $1000 price range. If you look inside, you'll see that these machines are typically mostly air inside. They could have been put in a package the size of a hardcover book, but consumers associate the big case with a powerful machine. Part of the reason these machines cost so much is profit-taking by the manufacturers, and part of it is the artificial impetus to get insanely powerful hardware, because software like Vista and OOo is coded so inefficiently. This whole setup is a house of cards, though. People don't need the equivalent of a 1990 supercomputer in order to send email and do their word-processing.
The trouble is that although a lot of small manufacturers have been testing the waters with lower-priced machines, the big ones haven't been interested. This is partly analogous to compact cars versus SUVs -- the profit margin on an SUV can be as much as $15,000, whereas the profit margin on a Ford Focus might be under $1000. Even if there's demand for the Focus, Ford has been more interested in pushing the SUV, because that's where the profit was. Then you have Apple selling a tightly integrated package of hardware and software, which people are willing to pay big premiums for. There's also the Windows tax, which hides the vast differences in hardware cost between a bleeding edge machine and something with lower specs.
For a long time, the only low-cost PCs I was ever able to find in retail outlets were the Great Quality PCs sold at Fry's, which came with Linux preinstalled. They were wonderful machines, and I still have a bunch of them in a lab at school, working great. They sold for about $200. However, Fry's stopped carrying them about a year ago. Apparently the high rate of returns was eliminating their profit margin. A lot of users were buying them to put pirated copies of Windows on, and then if they had a problem with the install, they'd return the machine.
There's also the Everex gPC. I own one, and reviewed it. Perfectly reasonable hardware, although the linux distro they put on it was junk. Judging from the customer feedback on WalMart's site, they've been some of the same problems as Great Quality with keeping their gPC customers satisfied -- a lot of people buying them apparently don't understand that the machine they're buying doesn't do Microsoft.
It's great to see something like the CherryPal come out. One interesting thing about it is that they're exploring the low end of the hardware specs that are necessary to run a web browser. This is conceivably a way for them to get around the low profit margins that have so far crippled investment in this end of the market. Here's a comparison of the specs of three cheap consumer linux boxes:
The Linksys v. 4 router cost something like $50 when it was available. (Later versions downgraded the specs and used a different OS instead of Linux.) Let's estimate what it would have cost today to upgrade its specs to something more like a desktop system (assuming it had been an upgradable system, which it wasn't). Paying retail today it would cost me $45 for a 1.8 GHz celeron cpu, $23 for 512 MB of ram, and $15 for a 4 GB keychain drive. Adding that on to the $50 retail price of the router, you get $133. Of course a computer manufacturer wouldn't be paying anything like these retail prices for the parts, so this is really a vast overestimate of what it would cost to manufacture a system like the CherryPal. I suspect their manufacturing price is more like $50.
Residential solar installations typically have no batteries, so there is no maintenance cost for batteries, nor replacement costs.
Just to expand a little on what you said, non-grid-tied systems with batteries do exist, but they're typically only used in remote rural areas, where the grid doesn't reach. You'd be insane to install one in any area where the grid is available.
I agree that the S&P's return is a riskier portfolio, and hindsight is 20-20. However, to get a simpler approach, just compare it to your mortgage at say 7%. Now the decision to put the money into your mortgage vs. your rooftop is "safe as houses."
My mortgage is 5%, and the interest tax-deductible.
Take away the rebate and it's 37 cents/kwh for his system, more expensive than any power anywhere.
I hear people bring up the rebates a lot as an argument against photovoltaics. It doesn't make sense. First of all, as an individual, my cost is my cost. If the rebate reduces my cost, then that reduced cost is the cost I need to consider. If, on the other hand, you want to talk about public policy, then you should compare apples and apples. Fossil fuels are massively subsidized by the government. The US has fought three wars in the Middle East since 1991. None of those would have been fought if there hadn't been oil in the Middle East. My grandkids are going to be paying the bills for the incredible budget deficits we've incurred because of these wars, and that's all a subsidy for fossil fuels, which are artificially cheap in the US compared to the rest of the world. Allowing people to burn fossil fuels and put CO2 in the atmosphere is another artificial subsidy; if they were paying the true economic costs of the greenhouse gases, the costs would be much higher. The interstate highway system is yet another gigantic federal subsidy for fossil fuels.
He says he put in $36,000 and will save $3,300 per year in payments to the power company. Now the historical annual rate of return of an S&P 500 index fund is 11.3% over the last century, so $36K put there would return over $4,000 -- enough to pay the $3,300 to the grid, have $700 left over and of course, still keeping the principal.
You're comparing apples and oranges here. If people always invested solely based on maximizing the expected average rate of return, then bank CDs wouldn't exist, bonds wouldn't exist, real estate and REITs wouldn't exist, etc. In reality, people are balancing the expected rate of return against variability. The reason bank CDs are paying something pathetic like 4% right now is that the banks have to guarantee that they'll pay you 4% on your money. The bank is essentially charging you for insurance against negative variability. Although there are a lot of unknowns involved in buying PV (what will electric rates do in the future? how much will the technology improve), they're a lot less than the unknowns involved in stocks. My S&P 500 index fund is down 19% from its peak value last year. There's a reason that most people build a balanced investment portfolio that includes both stocks and bonds; it's because mixing volatile and nonvolatile investments is a way of maximizes your expected rate of return for a given amount of risk that you're willing to accept.
And grid power might go up, but only so far. Because eventually the grid power hits the solar price, and the grid itself starts putting in solar sources at that price -- because it's cheaper.
Your logic doesn't work. When you own a PV system, you're part of "the grid itself." When I'm at work during the day, my PV panels are pumping energy into the grid, which is selling it to other customers. In my area, grid power has hit the solar price, for a south-facing roof with no shade; that's why I, a homeowner with a south-facing roof and no shade, have put in a PV system. You seem to be assuming that if rates go above a certain threshold, the entire state of California will suddenly magically cover itself with PV panels, because that will be the right thing to do according to the laws of supply and demand. That doesn't make sense, for a couple of reasons. First, there are huge variations in the price of land, the local cost of electricity, the amount of sunlight, which way people's roofs face, and how much shade they get. Second, there's a barrier to covering every house with PV panels, which is that most homeowners are short on capital. Your idea that the whole grid would suddenly go solar at some threshold is like imagining that everybody will suddenly drive a hydrogen-powered car if gas goes over $6 a gallon. We're talking about a massive infrastructure that doesn't change overnight.
By coincidence, today is the day I got my first yearly bill for my new photovoltaic system. Where I live (Orange County, CA, with Southern California Edison as my utility), people who have residential PV systems get billed yearly rather than monthly. A year is also pretty much the minimum amount of time for which you need data in order to find out how your system is performing, since both your energy production and your energy use fluctuate seasonally.
My bill for this year was $353.63. The system is nominally 4.4 kW, and cost $28k after rebate. It's covering about 90% of our use, which was almost exactly what we shot for -- if we produce more than we use over 12 months, they don't pay us for the excess.
People always want to know the number of years until the system pays for itself. Basically that's utterly impossible to predict. There's a reason that they exclude energy from the consumer price index -- it's because energy prices are extremely volatile. If the increased price of fossil fuels starts to be reflected in the cost of electricity, then I'm going to look like a financial genius. The other thing that's completely unknowable is how fast the technology will progress. If there's a breakthrough in technology five years from now, and the price of panels per kilowatt comes down by a factor of two, then I'll wish I'd waited. It's also kind of funny hearing the quick-buck psychological attitude a lot of Americans have toward investing money in something like this; from the way people talk, you'd think they were going to take that money that could have gone into photovoltaics and invest it in some kind of magical pixie dust that was guaranteed to pay a steady 20% annually until the end of time. And finally, beware of anyone making blanket statements about whether PV is ready for prime time or not. It completely depends on factors like the price of electricity in your area, which way your roof faces, your latitude, the amount of cloudy weather, and the amount of shade. PV is like Linux: it's ready for prime time for some people, and it's not ready for prime time for other people.
Linux seems to be a great system for people who want to burrow into the guts of the OS and make it do exactly what they want it to do. For people who are mostly just interested in using a computer as a tool to do things that are mostly unrelated to computers, it seems most of its flexibility and capability is wasted.
I think your comment would have been more valid about four years ago. Ubuntu is so good these days that you basically don't need to get into the "guts" at all. It just works. (Exceptions: power management for laptops, sharing a printer.)
To me, the reason for preferring linux has very little to do with flexibility or capability. The main reason I like it better is that I can run hundreds of applications without paying any money, and without having any of the hassles that come with buying and owning commercial software (drawers full of disks and licenses, being forced to upgrade in order to fix a bug or run the app on a newer OS, inability to try the software without buying it).
Other advantages: I got my kids linux boxes for under $200, which just isn't possible for Windows. I don't have to worry about viruses like I would on Windows. Removing applications completely is easy to do (unlike Windows, where people can't get IE off their machines, or can't find any way to uninstall the AV software they got on a one-month trial, or can't figure out how to deinstall the crapware that came on the machine). I don't have to keep paying $60 every six months or a year for an OS upgrade like Mac users do.
The only thing wrong with linux is lack of availability of 3rd party shrink-wrap type applications and games. I would love to give up XP, but linux can't run the video editing software that I need and games that I want.
I use linux as my only OS. I don't agree that lack of shinkwrapped apps is the only problem. In fact, I don't perceive the lack of shinkwrapped apps as a problem at all. I'm not into gaming, and as far as the rest of the software that I see on the aisles of retail stores, my usual thought when I look at it is My god, these poor shmucks pay $60 for this? I have software on linux that does the same thing, and it's free.
I'm not saying that you're wrong to want to run games. I'm just saying that your perception of that as the main problem is valid for some people like you, and not valid for some other people like me.
Here's my list of the main things that I perceive as problems with linux:
Power management has been difficult or impossible to get working properly on any machine I've ever tried it on. For a laptop, this is a real showstopper.
Sharing a printer is a disaster. I've spent endless hours trying to get this to work on my home network. Had it working once, a couple of years ago. Can't get it working anymore after upgrading ubuntu. I can't figure out why it should be so difficult to get one linux box to share its printer with another linux box.
Java applets are another thing that I can no longer get to work on my x64 box after an ubuntu upgrade. (This will probably get fixed soon.)
To my taste, the linux scheduler seems too oriented toward server use rather than desktop use. I would like it to have soft real-time, or to at least be more responsive under heavy I/O load. I know there are real-time patches for the kernel, but I don't want to have to do that level of voodoo to keep my system from locking up under certain circumstances.
OOo sucks, I'm convinced that OOo isn't going to stop sucking any time soon, and the alternatives to OOo also lack the combination of features, performance, and usability that I would consider decent.
Okay, I'm not trying to say your desires are the wrong desires to have, your experience is invalid, or anything like that. But re a few of your points:
Random crashes. I mean, probably as many or more as I get regularly in Windows, with the added inconvenience of ctrl+alt+bckspce not being near as good as ctrl+alt+delete, which brings up a handy task menu for me to clean up (usually).
This isn't very informative. What's crashing? A GUI app? Your WM? X? A binary blob inside the kernel? The kernel itself? For me, firefox crashes when I try to play youtube videos on my x64 ubuntu box; that's it -- I never experience any crashes on linux other than that. I'm not trying to say you're not experiencing crashes, just that it may not be universal, and it's not clear from your post what kind of crashes you're talking about or whether they have anything to do with linux per se.
Civilization 2. Best/funnest version of the game, will not play in wine even though it's like 10 years old.
Just like you, I love civ 2. But that doesn't mean I expect it to run on an OS it was never intended to run on.
I like how Windows arranges it's GUI, start button, quicklaunch, then task list, then systray and clock. Less real estate, all the same functionality, but without a top AND bottom bar.
So pick a different WM, one that works more like what you want. Personally, I don't want a top bar, and I don't want a bottom bar either, and that's why I use fluxbox. This is a huge argument in favor of linux, not against it. On Windows, you don't get a choice of WM.
I just got an email from a google developer working on google docs who saw my comment on above, and was happy that someone had noticed the improved performance. He says they started a feature freeze a few months ago, and spent the last few months working exclusively on performance.
I would think it would be impractical to put enough small but powerful electromagnets behind the fluid so that you could make a smooth surface.
Their paper says they surrounded the mirror with a Maxwell coil, which has an extremely uniform magnetic field on its interior.
Would it really be possible enough to make the magnetic field smooth enough so that the mirror surface was smooth and not something like the surface of a 300 sided polyhedron?
I think the idea is that they spin it so that it assumes a parabololoidal shape. A paraboloid is the mathematically perfect shape for bringing parallel rays to a focus. The magnetic fields are only used for small corrections to the shape. These small corrections might, for example, be used for adaptive optics, to correct for atmospheric turbulence on a real-time basis.
The slashdot summary talks about pointing away from the vertical, whereas the paper doesn't talk about that explicitly. I may be wrong, but I think the idea is this. A liquid metal telescope can only view a certain circular strip of the sky, which depends on the latitude at which it's located. You want this strip to be as wide as possible. Theoretically, you can just move your CCD (or whatever instrument it is) off the axis, and it will get a field of view that's away from vertical. However, any optical device is subject to aberrations if you try to use it far off axis. Reading between the lines here, I think the idea is that you can correct for the aberrations using the magnets, so that it might be possible to get good-quality images very far off the vertical axis -- "very far" meaning, I dunno, maybe five or ten degrees or something.
You might just be using a different browser with a better JavaScript interpreter... both Safari and Firefox have made great strides with JavaScript performance.
That could be. I think I was using ff 2 last time I tried it. I'm using ff 3 now.
The last time I tried the google docs spreadsheet (maybe 6 months ago?), it was ridiculously slow. I was about to post here and point that out, but then I thought I ought to check how the performance was today, in case it had improved. Well, I don't have any real data, but my subjective impression is that they must have made vast improvements in its performance since the last time I tried it. It really seems fine now.
The question in my mind now is how many people are really going to want this.
A lot of users aren't going to use it, for the same reason they're still running IE5 -- they've always "had Microsoft," and they're not the kind of people who are interested in tinkering with software.
Serious users aren't going to use it, because it doesn't have the right features.
I'm not going to use it, because I'm getting along fine with gnumeric and ooo, and I see web apps as a free-as-in-beer solution that would be a step backward from free-as-in-speech.
I teach physics at a community college, and I have a bunch of linux boxes in the lab alongside the windows machines. The linux boxes only have Ooo, and the Windows boxes have both Ooo and Excel.
It's been interesting seeing how students react to being presented with a choice between Excel and Ooo. I actually have documentation in the lab manual for Ooo, and none for Excel. Nevertheless, the vast majority don't want to mess with Ooo. Even if they have never used a spreadsheet before in their life, Excel is a brand name they've heard, so that's what they gravitate toward.
Of the two usenet providers you linked to, both only have monthly plans. For the vast majority of usenet users, who never use binary groups at all, it's a much better deal to pay by the gigabyte. Astraweb, for instance, sold me 25 Gb for $10, and I expect that to last until I'm dead.
It's certainly true that there's a ton of spam on usenet recently that *claims* to be from gmail addresses. That doesn't mean it really is from gmail addresses. You can claim any address you want to claim when you post on usenet. My usenet posts claim to be from an email address in the comain lightSPAMandISmatterEVIL.com.
Maybe so, but I'd be surprised if someone could disguise the rest of the header info to point completely at Google when they weren't involved at all.
Why? Forging headers is a lot easier than solving captchas. Virtually all spam for the last 5 years or more has had forged headers.
It's certainly true that there's a ton of spam on usenet recently that *claims* to be from gmail addresses. That doesn't mean it really is from gmail addresses. You can claim any address you want to claim when you post on usenet. My usenet posts claim to be from an email address in the comain lightSPAMandISmatterEVIL.com.
The trend is that ISPs are dropping usenet access. My ISP dropped it last month -- not just to binary groups but to all groups. I ended up deciding to pay someone else for usenet access, but many people aren't going to bother (or don't realize how cheap it can be if all you want is text groups). Usenet is in danger of drying up and blowing away completely, and if google groups helps to keep it above critical mass by providing a web interface, then I think we ought to hail google as the savior of usenet.
Slashdot Mirror
I'm a community college physics teacher in the U.S., and I scrounge computer hardware to use in my lab classes. The school provides one Windows box per lab group, i.e., 7 computers for a class of 25 students. The trouble with that is, you get one student doing the graphs on the computer, and the rest of the group just sits there and watches. I've made a geekly hobby out of putting together decent Linux systems from garage sales, thrift shops, etc., to supplement what the school provides. It was interesting comparing the article with my own experiences back here in the developed world.
One thing to keep in mind is that the line between good and bad hardware is extremely fuzzy. I picked up an old 500 MHz e-Machines box recently at Good Will for $89, and with a $20 memory upgrade it makes a perfectly decent Linux machine, especially with a distro like xubuntu that's designed more for low-end hardware (xfce rather than gnome, abiword rather than OOo). Many people would have considered this machine too old to be useful, but it works fine for the application I need it for.
Similar deal with monitors. I actually find that cheap monitors are much, much harder to find than cheap computers. You don't see them much at thrift shops or swap meets, I guess because CRTs are heavy and bulky in relation to what you can sell them for. When I do get an old CRT, its mean time to failure is usually pretty darn short, probably 12-24 months. As far as I can tell, computer CRTs have a certain lifetime, and when you get your hands on a cheap one it's already near the end of that.
One thing that's absurd, when you view computers as potential solid waste, is the amount of air inside a tower case these days. On a low-end machine, the case can easily be 90% empty. It's the equivalent of going to McDonalds and having them serve you your little 99-cent hamburger in a styrofoam clamshell the size of a microwave oven. I'm hoping the Asus eee Box comes out soon, and Asus doesn't jack up the price. For $269, it could be a wonderful deal.
And by the way, if you're in Orange County, CA, and have a working monitor you're willing to donate, please email me at crowell08 at lightSPAMandISmatterEVIL dot com. I'll be more than happy to come and pick it up, and you can have the warm fuzzy feeling of knowing that you're helping me spread peace, love, and linux to my students.
My dispute with this line of reasoning is that we use an insignificant amount of oil for electricity generation purposes.
Where I live, in Southern California, I believe SCE generates our electricity from oil and natural gas. Oil comes from oilfields. Natural gas comes from oilfields and natural gas fields.
[...] coal and natural gas [...] receive some of the most marginal amounts of subsidy in the industry
The WSJ article you link to is talking about direct, overt government subsidies. Those aren't the subsidies I'm talking about.
A couple of things to keep in mind here:
Paper, printing, and binding are usually not a significant percentage of the cost of a textbook. Digital textbooks might have the advantage of cutting out the campus bookstore as a middleman, but I suspect that the publisher is not going to let the student have the money that would have gone to the store (typically about 27%). Another issue is returns, and the ability to buy and sell used copies, which probably wouldn't be possible.
I can see how it could come in handy if you're on vacation and want to travel light, but IMO that's not nearly enough to overcome the negatives. I'll probably get an e-book reader in 2030 or something. There's no rush. First I want to see someone get it right.
I don't think the situation isn't quite that rosy. Although SWF is an open format, there are lots and lots of other bits and pieces that you need, such as the library of GUI widgets, and the various audio and video codecs. Those are all proprietary. (Some, e.g., mp3, are only proprietary because someone else owns a patent. However, Adobe hasn't bothered to implement support for free codecs such as vorbis and theora.) It's true that there are third-party, open-source compilers such as haxe. However, there's nothing that's compatible at the source-code level with Adobe's tools. Therefore if you're trying to work on an existing flash project, or trying to learn flash development from books, you really have no option other than Adobe's proprietary tools. The last time I checked, their tools also came with nasty EULAs, which, e.g., forbid reverse-engineering, creation of competing products, or use on a mobile device.
Oh, please. You've done a good job of using impressive terms to sound like you know what you're talking about. If you want to talk about the real-world risks of having your crypto broken, then you need to consider all the real-world methods by which your crypto could be broken. It doesn't matter that a one-time pad can be theoretically proved to be invulnerable to certain kinds of attacks, to which various symmetric and asymmetric ciphers are theoretically vulnerable. What matters is the actual types of attacks that are practical and likely, and the actual problems you'll have in the practical implementation of a particular method. If you're using a one-time pad, then there are several obvious, well-known things that can go wrong: (1) you have to physically exchange the one-time pads, which may be difficult to do (and do securely) if the NSA is really following you everywhere, opening your mail, etc.; (2) both parties have to maintain the security of their own copies of the one-time pads, which may be difficult to do if the NSA is really determined to get them; (3) there is a tendency for users to get lazy and reuse a one-time pad, which then makes you vulnerable to certain kinds of attacks. Standard symmetric and asymmetric ciphers are more or less immune to these problems (#1: swapping passwords securely is a lot easier than swapping large amounts of binary data securely; #2: you can keep the password in your head instead of stuffing a keychain drive under your mattress; #3: no such issue). Yes, there are also certain kinds of attacks to which standard ciphers are vulnerable and one-time pads invulnerable (e.g., dictionary attacks on your password, shoulder-surfing,...) One-time pads are not magic pixie dust for cryptography. There is no magic pixie dust for cryptography. The good news is that we're living in a golden age of privacy, in the sense that you can legally, publicly get software to do encryption so good that essentially your main worry is no longer the encryption, it's the social/personal/legal issues surrounding its implementation.
Yeah. This also seems to be an example of a more general phenomenon with Linux support, which is that the same company will make completely contradictory statements about their own Linux support. In the earlier slashdot story, someone from Foxconn is directly quoted as saying 'it doesn't support Linux;' now they say they always intended to support Linux. The truth is probably that they never even thought about Linux support, and then when the issue was brought to their attention random representatives started saying random things off the cuff.
I've had a similar experience with Amazon's MP3 store. If you want to buy entire albums (as opposed to individual tracks), you have to use special downloading software that they supply. The software was initially only available in Windows and Mac versions, but pretty quickly they brought out Linux versions as well. Nowadays when you use your Linux box to shop for albumbs on their site, if you don't have the software installed your browser will detect that, and detect your OS as linux, and they'll generate a page for you offering links to download a linux version of the downloader. In fact, they even have it available in multiple versions for different linux distros. However, the linux downloader has been pretty buggy for me (and was also hard to get working properly on x64). I've had it working, then it broke, etc. I've done two calls to Amazon's tech support about this, and in both cases, the initial reaction was to tell me to do a bunch of stuff (with the usual confusion because the Indian tech support person gives Windows+IE instructions, and has never heard of Linux), and then when that didn't help they checked with someone else, who told them Linux wasn't supported. Never mind that they've had Linux versions of the software up on the site for months now.
I think part of the problem is that so many people in the hardware and software industries live in a 100%-Windows environment. It honestly never even occurs to them that anyone is running any other OS. (In the case of Foxconn, they're not making mac-compatible boards, so it's probably true that 99% of their boards are being used with Windows.) Then when the issue comes up, they just deal with it off the cuff. It's like asking them what their policy is on recycling cardboard -- they probably don't have one, and they don't see why it's important.
Another problem may be that in a Windows monoculture environment, many people don't understand what a standard really is. They think Windows and Word and IE are standards. Instead of developing for the relevant standard, some PHB makes the decision that they're going to target something proprietary, calling that a "standard," and they think of it as extra work to add support for anything else -- when in fact, it would have made more sense just to support the standard properly in the first place.
I agree that for a machine I want to work on myself, a tower is great. Heck, even with a tower there are times when I'm scraping my knucles and cussing and swearing. But when's the last time you upgraded your transistor radio? We're headed into an era when PCs will sell for $50, and will be disposable.
Sure, but that's only for those applications. You could compare it to a TV. A lot of affluent families these days own four TVs, but only one of those is the giant home-theater one. The rest are cheap consumer stuff from Costco, which they'll throw in a dumpster in 5 years.
The problem with this device is that it isn't that much cheaper than a full budget PC that will whack this into the ground.
What figure do you have in mind when you say "isn't that much cheaper?" It's basically impossible to walk into a retail store and buy a new PC for less than $400-500. The Everex gPC does have better specs, and is only $200, but it's only available online. For people who like the small form factor, the CherryPal might be preferable to the gPC. The gPC's Linux distro also sucks to high heaven, so for people who aren't technically adept enough to install a linux distro themselves, there really aren't any good options out there right now for less than $400-500.
I think we're currently in a period of marketplace chaos, and when the dust settles we'll find that a $1000 PC in a tower case seems about as archaic as a radio in a wooden case the size of a washing machine.
The biggest computer manufacturers are still selling machines in the $1000 price range. If you look inside, you'll see that these machines are typically mostly air inside. They could have been put in a package the size of a hardcover book, but consumers associate the big case with a powerful machine. Part of the reason these machines cost so much is profit-taking by the manufacturers, and part of it is the artificial impetus to get insanely powerful hardware, because software like Vista and OOo is coded so inefficiently. This whole setup is a house of cards, though. People don't need the equivalent of a 1990 supercomputer in order to send email and do their word-processing.
The trouble is that although a lot of small manufacturers have been testing the waters with lower-priced machines, the big ones haven't been interested. This is partly analogous to compact cars versus SUVs -- the profit margin on an SUV can be as much as $15,000, whereas the profit margin on a Ford Focus might be under $1000. Even if there's demand for the Focus, Ford has been more interested in pushing the SUV, because that's where the profit was. Then you have Apple selling a tightly integrated package of hardware and software, which people are willing to pay big premiums for. There's also the Windows tax, which hides the vast differences in hardware cost between a bleeding edge machine and something with lower specs.
For a long time, the only low-cost PCs I was ever able to find in retail outlets were the Great Quality PCs sold at Fry's, which came with Linux preinstalled. They were wonderful machines, and I still have a bunch of them in a lab at school, working great. They sold for about $200. However, Fry's stopped carrying them about a year ago. Apparently the high rate of returns was eliminating their profit margin. A lot of users were buying them to put pirated copies of Windows on, and then if they had a problem with the install, they'd return the machine.
There's also the Everex gPC. I own one, and reviewed it. Perfectly reasonable hardware, although the linux distro they put on it was junk. Judging from the customer feedback on WalMart's site, they've been some of the same problems as Great Quality with keeping their gPC customers satisfied -- a lot of people buying them apparently don't understand that the machine they're buying doesn't do Microsoft.
It's great to see something like the CherryPal come out. One interesting thing about it is that they're exploring the low end of the hardware specs that are necessary to run a web browser. This is conceivably a way for them to get around the low profit margins that have so far crippled investment in this end of the market. Here's a comparison of the specs of three cheap consumer linux boxes:
The Linksys v. 4 router cost something like $50 when it was available. (Later versions downgraded the specs and used a different OS instead of Linux.) Let's estimate what it would have cost today to upgrade its specs to something more like a desktop system (assuming it had been an upgradable system, which it wasn't). Paying retail today it would cost me $45 for a 1.8 GHz celeron cpu, $23 for 512 MB of ram, and $15 for a 4 GB keychain drive. Adding that on to the $50 retail price of the router, you get $133. Of course a computer manufacturer wouldn't be paying anything like these retail prices for the parts, so this is really a vast overestimate of what it would cost to manufacture a system like the CherryPal. I suspect their manufacturing price is more like $50.
Residential solar installations typically have no batteries, so there is no maintenance cost for batteries, nor replacement costs.
Just to expand a little on what you said, non-grid-tied systems with batteries do exist, but they're typically only used in remote rural areas, where the grid doesn't reach. You'd be insane to install one in any area where the grid is available.
I agree that the S&P's return is a riskier portfolio, and hindsight is 20-20. However, to get a simpler approach, just compare it to your mortgage at say 7%. Now the decision to put the money into your mortgage vs. your rooftop is "safe as houses."
My mortgage is 5%, and the interest tax-deductible.
Take away the rebate and it's 37 cents/kwh for his system, more expensive than any power anywhere.
I hear people bring up the rebates a lot as an argument against photovoltaics. It doesn't make sense. First of all, as an individual, my cost is my cost. If the rebate reduces my cost, then that reduced cost is the cost I need to consider. If, on the other hand, you want to talk about public policy, then you should compare apples and apples. Fossil fuels are massively subsidized by the government. The US has fought three wars in the Middle East since 1991. None of those would have been fought if there hadn't been oil in the Middle East. My grandkids are going to be paying the bills for the incredible budget deficits we've incurred because of these wars, and that's all a subsidy for fossil fuels, which are artificially cheap in the US compared to the rest of the world. Allowing people to burn fossil fuels and put CO2 in the atmosphere is another artificial subsidy; if they were paying the true economic costs of the greenhouse gases, the costs would be much higher. The interstate highway system is yet another gigantic federal subsidy for fossil fuels.
He says he put in $36,000 and will save $3,300 per year in payments to the power company. Now the historical annual rate of return of an S&P 500 index fund is 11.3% over the last century, so $36K put there would return over $4,000 -- enough to pay the $3,300 to the grid, have $700 left over and of course, still keeping the principal.
You're comparing apples and oranges here. If people always invested solely based on maximizing the expected average rate of return, then bank CDs wouldn't exist, bonds wouldn't exist, real estate and REITs wouldn't exist, etc. In reality, people are balancing the expected rate of return against variability. The reason bank CDs are paying something pathetic like 4% right now is that the banks have to guarantee that they'll pay you 4% on your money. The bank is essentially charging you for insurance against negative variability. Although there are a lot of unknowns involved in buying PV (what will electric rates do in the future? how much will the technology improve), they're a lot less than the unknowns involved in stocks. My S&P 500 index fund is down 19% from its peak value last year. There's a reason that most people build a balanced investment portfolio that includes both stocks and bonds; it's because mixing volatile and nonvolatile investments is a way of maximizes your expected rate of return for a given amount of risk that you're willing to accept.
And grid power might go up, but only so far. Because eventually the grid power hits the solar price, and the grid itself starts putting in solar sources at that price -- because it's cheaper.
Your logic doesn't work. When you own a PV system, you're part of "the grid itself." When I'm at work during the day, my PV panels are pumping energy into the grid, which is selling it to other customers. In my area, grid power has hit the solar price, for a south-facing roof with no shade; that's why I, a homeowner with a south-facing roof and no shade, have put in a PV system. You seem to be assuming that if rates go above a certain threshold, the entire state of California will suddenly magically cover itself with PV panels, because that will be the right thing to do according to the laws of supply and demand. That doesn't make sense, for a couple of reasons. First, there are huge variations in the price of land, the local cost of electricity, the amount of sunlight, which way people's roofs face, and how much shade they get. Second, there's a barrier to covering every house with PV panels, which is that most homeowners are short on capital. Your idea that the whole grid would suddenly go solar at some threshold is like imagining that everybody will suddenly drive a hydrogen-powered car if gas goes over $6 a gallon. We're talking about a massive infrastructure that doesn't change overnight.
By coincidence, today is the day I got my first yearly bill for my new photovoltaic system. Where I live (Orange County, CA, with Southern California Edison as my utility), people who have residential PV systems get billed yearly rather than monthly. A year is also pretty much the minimum amount of time for which you need data in order to find out how your system is performing, since both your energy production and your energy use fluctuate seasonally.
My bill for this year was $353.63. The system is nominally 4.4 kW, and cost $28k after rebate. It's covering about 90% of our use, which was almost exactly what we shot for -- if we produce more than we use over 12 months, they don't pay us for the excess.
People always want to know the number of years until the system pays for itself. Basically that's utterly impossible to predict. There's a reason that they exclude energy from the consumer price index -- it's because energy prices are extremely volatile. If the increased price of fossil fuels starts to be reflected in the cost of electricity, then I'm going to look like a financial genius. The other thing that's completely unknowable is how fast the technology will progress. If there's a breakthrough in technology five years from now, and the price of panels per kilowatt comes down by a factor of two, then I'll wish I'd waited. It's also kind of funny hearing the quick-buck psychological attitude a lot of Americans have toward investing money in something like this; from the way people talk, you'd think they were going to take that money that could have gone into photovoltaics and invest it in some kind of magical pixie dust that was guaranteed to pay a steady 20% annually until the end of time. And finally, beware of anyone making blanket statements about whether PV is ready for prime time or not. It completely depends on factors like the price of electricity in your area, which way your roof faces, your latitude, the amount of cloudy weather, and the amount of shade. PV is like Linux: it's ready for prime time for some people, and it's not ready for prime time for other people.
I think your comment would have been more valid about four years ago. Ubuntu is so good these days that you basically don't need to get into the "guts" at all. It just works. (Exceptions: power management for laptops, sharing a printer.)
To me, the reason for preferring linux has very little to do with flexibility or capability. The main reason I like it better is that I can run hundreds of applications without paying any money, and without having any of the hassles that come with buying and owning commercial software (drawers full of disks and licenses, being forced to upgrade in order to fix a bug or run the app on a newer OS, inability to try the software without buying it).
Other advantages: I got my kids linux boxes for under $200, which just isn't possible for Windows. I don't have to worry about viruses like I would on Windows. Removing applications completely is easy to do (unlike Windows, where people can't get IE off their machines, or can't find any way to uninstall the AV software they got on a one-month trial, or can't figure out how to deinstall the crapware that came on the machine). I don't have to keep paying $60 every six months or a year for an OS upgrade like Mac users do.
I use linux as my only OS. I don't agree that lack of shinkwrapped apps is the only problem. In fact, I don't perceive the lack of shinkwrapped apps as a problem at all. I'm not into gaming, and as far as the rest of the software that I see on the aisles of retail stores, my usual thought when I look at it is My god, these poor shmucks pay $60 for this? I have software on linux that does the same thing, and it's free. I'm not saying that you're wrong to want to run games. I'm just saying that your perception of that as the main problem is valid for some people like you, and not valid for some other people like me.
Here's my list of the main things that I perceive as problems with linux:
Okay, I'm not trying to say your desires are the wrong desires to have, your experience is invalid, or anything like that. But re a few of your points:
This isn't very informative. What's crashing? A GUI app? Your WM? X? A binary blob inside the kernel? The kernel itself? For me, firefox crashes when I try to play youtube videos on my x64 ubuntu box; that's it -- I never experience any crashes on linux other than that. I'm not trying to say you're not experiencing crashes, just that it may not be universal, and it's not clear from your post what kind of crashes you're talking about or whether they have anything to do with linux per se.
Just like you, I love civ 2. But that doesn't mean I expect it to run on an OS it was never intended to run on.
So pick a different WM, one that works more like what you want. Personally, I don't want a top bar, and I don't want a bottom bar either, and that's why I use fluxbox. This is a huge argument in favor of linux, not against it. On Windows, you don't get a choice of WM.
I just got an email from a google developer working on google docs who saw my comment on above, and was happy that someone had noticed the improved performance. He says they started a feature freeze a few months ago, and spent the last few months working exclusively on performance.
Their paper says they surrounded the mirror with a Maxwell coil, which has an extremely uniform magnetic field on its interior.
I think the idea is that they spin it so that it assumes a parabololoidal shape. A paraboloid is the mathematically perfect shape for bringing parallel rays to a focus. The magnetic fields are only used for small corrections to the shape. These small corrections might, for example, be used for adaptive optics, to correct for atmospheric turbulence on a real-time basis.
The slashdot summary talks about pointing away from the vertical, whereas the paper doesn't talk about that explicitly. I may be wrong, but I think the idea is this. A liquid metal telescope can only view a certain circular strip of the sky, which depends on the latitude at which it's located. You want this strip to be as wide as possible. Theoretically, you can just move your CCD (or whatever instrument it is) off the axis, and it will get a field of view that's away from vertical. However, any optical device is subject to aberrations if you try to use it far off axis. Reading between the lines here, I think the idea is that you can correct for the aberrations using the magnets, so that it might be possible to get good-quality images very far off the vertical axis -- "very far" meaning, I dunno, maybe five or ten degrees or something.
That could be. I think I was using ff 2 last time I tried it. I'm using ff 3 now.
The last time I tried the google docs spreadsheet (maybe 6 months ago?), it was ridiculously slow. I was about to post here and point that out, but then I thought I ought to check how the performance was today, in case it had improved. Well, I don't have any real data, but my subjective impression is that they must have made vast improvements in its performance since the last time I tried it. It really seems fine now.
The question in my mind now is how many people are really going to want this.
I teach physics at a community college, and I have a bunch of linux boxes in the lab alongside the windows machines. The linux boxes only have Ooo, and the Windows boxes have both Ooo and Excel. It's been interesting seeing how students react to being presented with a choice between Excel and Ooo. I actually have documentation in the lab manual for Ooo, and none for Excel. Nevertheless, the vast majority don't want to mess with Ooo. Even if they have never used a spreadsheet before in their life, Excel is a brand name they've heard, so that's what they gravitate toward.
Of the two usenet providers you linked to, both only have monthly plans. For the vast majority of usenet users, who never use binary groups at all, it's a much better deal to pay by the gigabyte. Astraweb, for instance, sold me 25 Gb for $10, and I expect that to last until I'm dead.
Why? Forging headers is a lot easier than solving captchas. Virtually all spam for the last 5 years or more has had forged headers.
It's certainly true that there's a ton of spam on usenet recently that *claims* to be from gmail addresses. That doesn't mean it really is from gmail addresses. You can claim any address you want to claim when you post on usenet. My usenet posts claim to be from an email address in the comain lightSPAMandISmatterEVIL.com.
The trend is that ISPs are dropping usenet access. My ISP dropped it last month -- not just to binary groups but to all groups. I ended up deciding to pay someone else for usenet access, but many people aren't going to bother (or don't realize how cheap it can be if all you want is text groups). Usenet is in danger of drying up and blowing away completely, and if google groups helps to keep it above critical mass by providing a web interface, then I think we ought to hail google as the savior of usenet.