California is renowned mostly because it's pretty much the only area in the USA that produces acceptable wine, not because that wine is particularly good in comparison to the rest of the world, just because it's the best the USA can produce.
Not to kill too many sacred cows here, but the French disagree with you rather soundly. According to them, California wines are the finest in the world. I'm partial to Australian wines myself, but then I've got family down under, which does influence me a bit.
Personally, I think that most "wine experts" are overblown windbags who engage in the worst overuse of metaphors in modern language. However, I do happen to agree with them now and again: every single time I've had anything from Stag's Leap, it's been incredible. Wines, like many other issues of taste, are difficult to come to any lasting consensus. You would do well to continue to trust your own taste over anyone else's.
(If you're in the mood for advice, you would do yourself a favor to lose the bit about California wines not being world class. You just come across as uninformed.)
I have enjoyed chocolate around the world, including Belgium and Switzerland and IMHO, craft made chocolates from the US (I'm specifically thinking of Sharffen Berger (Hershey), Ghirardelli (Lindt), and Dove (Mars)) are the equal of any chocolatier anywhere in the world.
Also, Cadbury and Kraft are in the process of ruining many European chocolate brands, so it's not like the rest of the world is free from crap chocolate.
Of course, my palate may not be as refined as yours, so I'm not expecting universal agreement on such a personal subject as chocolate. But there is excellent US-made chocolate. It is simply necessary to be somewhat discriminating.
They never sound right so it will never be cool enough.
I think silence and/or natural noises (wind noise) are pretty damned cool. But then, I prefer sailboats over motor boats, vibrating phones over polyphonic ring tones, opening the window over central heating/AC, backpacking over theme parks, reading over television...
So, I'm a wierdo. But I did manage to find a wife who agrees with me on noise, so I'm not alone, just outnumbered.
Less glibly: I would love to be able to eliminate my motorcycle tailpipe and make it completely silent. I've heard that this would make me less safe, but I've noticed that when driving, I've never heard a motorcycle coming up behind me. Even the ones with loud pipes.
But he did not copy the code. Rather it was just similar.
Erm, no. The developer admitted that he copied the code over so that he could make progress on another part of the code. It was his intention to remove the copied code at a later date. By not even leaving any notes in the code about the copy and checking that code into a public cvs repository, the developer made a pretty big (yet fixable) mistake.
I'll admit it, I've done that very thing (copy some code in order to bootstrap a project). But I put all sorts of comments around tainted code, and I make damned sure that every single line of tainted code is rewritten before that code, or a product based on it, is released into the wild.
Theo and Marcus both come across as graceless and petulant children. At least Marcus decided to be childish quietly. Theo's ongoing rants about "the inhumanity" of it all just get hysterical after a few posts. Yes, the original email probably should have been more private. But the response from Theo is completely and utterly over the top.
Collection of random thoughts in response to your post:
You're not thinking about the costs on an apples to apples basis. If you've decided to rebuild a car, how much does a new (remanufactured) drivetrain cost? Having just finished a frame-off Jeep Scrambler buildup with: a rebuilt chevy 350 engine through a new centerforce clutch into a rebuilt NV4500 5-speed and then into a new Atlas transfer case... all-up cost of the powertrain is about $12k. Pretty comparable, given that one is for extreme performance/durability and the other is for extreme efficiency.
What you're really paying for there is an EV drivetrain that doesn't throw away batteries. The solectria controller should give you very good life on your batteries. Chewing through batteries is a common medium term problem with EV converisons, and I have no intention of spending that much money over and over again.
You definitely start from a salvage car to minimize the cost of the whole project. There are cheap 914's to be had on eBay (mostly partially completed project cars). The original engines in those things sucked.
In the EV, I'm taking the time to refurbish/replace/rebuild as much of the drivetrain as I can to guarantee that the car will last as long as a "new" car, and therefore be worth the $16k total cost of the project. BTW, I'm using the "AC heavy manual transmission" kit from them, which is also $10k.
Still, I need a donor vehicle first, and my minimum range would be around 120 miles, or at least 80 at highway velocities. I know full well that this would double the cost of my battery pack versus yours.
These guys sell a kit for getting that kind of range in a Porsche 914 but I wanted more utility and I don't require quite that much range to consider the result a very useful vehicle.
30 minute average for 7 miles? That's only a 14mph average. You have to traverse roads with speeds in excess of 40mph? No redlight?
*scratches head* SoCal is messed up.
You have identified the problem correctly. My commute has an enormous number of stops and between the stops, everyone's dashing like mad. Dashing when they aren't completely stopped in traffic, that is.
The big issue with my commute is that my home and my office are on opposite sides of UCLA. UCLA has a huge commuter student population and is usually surrounded with nasty traffic on roads with 35mph signs that everyone ignores, choosing instead to go 50mph when they can. These roads also lack any sort of bicycle lane and usually have no parking lane or shoulder. If I stay away from UCLA I can travel on slower/safer roads, but going that far around UCLA means a much longer trip.
With the motorcycle, I can slide through traffic and then go as fast or faster as the other cars when traffic speeds up. The bicycle is even better at slipping through traffic, but when things speed up on these roads, it's pretty dangerous if you can't keep up.
From what I've read, 3k is for a HV, not an EV. EV packs have to be substantially larger. And 6 cents/mile is what a efficient compact car uses per mile in gasoline, and that's not an up front price, so it's effectivly cheaper.
20 x $160 is the pack for this car. Less the core charge, it comes to right around $3000. And given local gas prices (>$3/gal), I suspect that compact cars cost more per mile around here.
With only seven miles for your commute, you'd be better off in most situations with a bicycle, or at most one of those overgrown golf carts.
Possible, and I looked at that, but the roads that I have to traverse have typical speeds of 45-50mph, and there are no alternate/slower routes that don't also more than double the distance.
At the moment, I ride a motorcycle and spend $12 every two weeks on gas to travel ~160 miles, which comes to about $0.08/mile. I agree that the EV will cost more than the motorcycle per mile, and that's been a question about this project from the beginning. The other issue is that I like riding the motorcycle and getting to work that much faster (lane splitting is legal in CA, so I spend 15 minutes instead of the 25-35 minutes it take on those days when I take the car for whatever reason).
Being really honest, at least half of the fun is the project itself.
most people would need a second car for when the distance limitations become a problem.
Car rental may be more practical for the occasional long trip.
how does the distance fall off over time as the batteries age?
It depends on how the batteries degrade. Most of the time, you'll have 75% of the base capacity when you replace the batteries on schedule.
In that $3.50, do you factor in the periodic replacement batteries? How many charges can you get on a single set of batteries, and how many batteries in the set?
No, that number is the cost of the electricity only at $0.14/kWH. With my usage profile (my round-trip commute uses less than 1/3 of a charge), I should be able to get at least 50,000 miles or about 1000-1200 partial charge cycles. This vehicle uses twenty $160 12V deep cycle batteries. With the core charge, a new pack will cost about $3000. So the batteries cost about $0.06/mile or slightly more than the electricity.
Also, you specify $15k to do the conversion to electric.
It's actually closer to $16k, but that's for everything, including the salvage car, bodywork, repainting, overhaul of the manual transmission, the electric conversion (with my labor), and the first battery pack. The batteries are maintenance items, and every five years or 50,000 miles isn't a bad schedule for replacement. There are two extra pumps (power brakes, power steering) that will require periodic maintenance/replacement as well, and then some of the usual car items: brake pads, rotors, suspension gear, door and window seals, heating and A/C, etc.
And recycling is more complicated than it first looks. It's entirely possible that the recycling process is safe, recovers more chemicals than it takes to remove them, doesn't require huge amounts of fossil fuels be burned in recovery, but it's also possible that the opposite is true.
Lead-acid recycling is pretty well solved. Obviously, more can be done, but the work will be improving a successful process instead of trying to come up with a new way to solve a big problem.
Gee, how much energy did I say an EV system had compared to a gallon of gas?
I was objecting to the "with similar range" part of your statement because you're right about the quantity of energy. The problem I had is that EV's are much more efficient about converting chemical energy into range: even though I can only carry about 29kWH of usable energy (about a half gallon of gas equivalent, which seems about right), I will get about the same range as three gallons of gas.
The 85 mile range is stated as conservative based on the very similar cars I'm using as sources for my design. One is a '93 Saturn SL2 (almost exactly the same curb weight), and when he drives carefully, he can get 90 miles before the controller begins complaining, giving him five more miles to get home with his setup. I'm using a higher efficiency motor than him, but also one fewer 12V batteries, so 85-90 miles per charge doesn't seem particularly ambitious.
The reason is that a 10h charge gets you 1 hour on the road.
In the EV I'm converting (1998 Saturn SW2 wagon), twelve hours of charging (110V/20A) gets me 85 miles. That's about 25kWh or $3.50. Since the car used to get 28mpg, that's about a third the price of gas for the same distance, even with SoCal's inflated electricity prices (we also have inflated gas prices). For everything up to the occasional road trip, 85 miles per day is more than enough. We've got a second car for the road trips, so that's covered too.
If you want an electric car, you're probably going to have to deal with renting the battery, and having swap-out stations.
Unlikely, at best. Manufacturers will have to come to some agreement on what a standard pack looks like, mechanical and electrical connections, etc. Can't have too much variety (or the storage cost goes through the roof), but you'll have to support commuter cars up through heavy trucks. A battery swap station will need a large warehouse built on good foundations to store enough battery packs to service an average day. And why would I want to exchange my carefully maintained batteries that cost me $3000 with an expected life of 100k miles for some unknown set that someone else may have abused or be near the end of life?
Not to say what you're talking about will never happen (though I'm extremely skeptical), but for 95%+ of my driving, plugging in overnight works great. Actually, since my commute is only 7 miles each way, I'll only need to charge for about two hours to get back to 80%+. Now, if I need to drop the kids off at a couple of activities, get groceries, do a few other errands, and take the family out for an evening's entertainment? The car will need to charge the whole time that I'm asleep and be topped off with a full charge when I wake up in the morning.
Electric cars are a lot more practical than most people think. Mine will cost me $15k, take 200 hours of my time, will do 80mph and get 85 miles on a charge that costs me $3.50 while hauling four people or light hauling. Some might want to include the 200 hours in the cost, however, I won't bill myself for the time since I find it so enjoyable to work on it and would pay extra to have this much fun:)
On top of what you've said, I'd like to point out that suburbia in and of itself isn't the problem either. I'm personally getting impatient for the idiots that run this county to finally run a viable light rail system, but I'm worried that people won't use it.
Are you kidding? Viable? Not going to happen. LA installed a multi-line train from various places in and out of downtown. One of the places that the train might go: LAX airport. But no. The train stops two miles from the airport where if you take the train, you get to wait for a bus to take you the rest of the way to the terminal.
The original plan for the train had three stops inside the U-shaped terminal, and there's even a sub-level of the LAX buildings that was originally designed to accomodate the eventual train system. So what happened? The taxi drivers said that if the train was in the terminal, their business would be seriously hurt. So the train stops two miles from the terminal, and you get to stand outside and wait for a bus to take you the rest of the way. There are simply too many political interests with the clout to make sure that public officials don't threaten their business, even if they are part of the real problem.
I don't travel that often, but when I do, it's a rather big PITA to figure the best way to get to the airport. Guess why I never take the train. Because the train sucks, at the request of the taxi drivers? You got it in one! Yeah, I'm only a little bitter about that.
Flooded lead-acid batteries in a well-designed EV will last between 50k-100k miles (lots of variables there). Perhaps you think that's pitifully short, but I'm not so dissapointed in that number. A new lead-acid traction pack will cost about $3000, so that's about $0.03-$0.06/mile spent on the batteries.
God help you if you run out of batteries and don't get around to recharging in time, there's a few hundred kilograms of toxic material to dispose of and a few dozen thousands of dollars to replace them
Are you referring to damaging batteries through overuse? Any modern EV system controller will keep batteries above 20% charge (much cheaper to get a tow after bad planning than to damage the traction pack). As for recyclability, flooded lead-acid batteries have a near 100% recylability and you'll trade then in for the core charge when you buy the replacements. Or NiMH batteries, from which the valuable nickel is recycled into stainless steel, if your community has a battery recycling program (mine does, here in SoCal). To be honest, though, nobody doing their own conversion will use NiMH cause big NiMH batteries are simply too expensive.
and an energy density not high enough to give any sort of range (in an EV you're lucky if you can fit about a gallon of gas worth of electrical energy onboard, with similar range).
Sorry, but that's bunk. I'm in the process of converting a 1998 Saturn SW2 station wagon to an EV and according to the system specs I'll conservatively get between 85-90 miles per charge with a (full charge: 12 hours of 20A @ 110V or about $4 in SoCal). Since my commute is 7 miles one way, 85 miles/day leaves lots of range left over for errands, rides for kids, short trips (to places with plugs), whatever. And we've got the Jeep with the 30 gallon tank (only 20 mpg FWIW) for longer trips.
I'd say to try again with the numbers, perhaps with a vehicle more suitable for EV conversion. Starting from a compact car, EV's are MUCH more practical than what you're talking about.
You've inspired me to do some more research into the possibilities of a low-power desktop. After all of this research? A 2-3 year old used laptop with the screen powered down (or a 1-2 year old Mac mini) seems to be the best bet for a low-power 24/7 server that doesn't break the bank on initial cost or power bills over time.
Slightly lower than you can from even the best laptops.
This is an article by Silent PC Review on a "Mobile on the Desktop" build up, and the best they were able to do was 22W/57W. They did use a Core 2 mobile CPU that will smoke anything I've ever considered building, but... When all is said and done, however, their build has a huge number of compromises to get "close, but not quite" to what all of my laptops use in the way of power.
If you know of more information on the web similar to http://www.spcr.com/ that would help me learn more on this subject, please post it. I'm definitely interested!
I said, in no uncertain terms, "you can build a desktop more efficiently."
I'm also talking about building a low-power computer, and in my experience, I haven't been able to get close to laptop power usage, let alone better than laptop power usage.
Slightly lower than you can from even the best laptops. Mobile components work just as well in desktops as they do in laptops (eg Turion CPU). However, you can get slightly higher efficiency from the power conversion, slightly more efficient components like video. Eliminate fans, and other benefits like lower price.
I just finished building a low-power HTPC with an Antec NSK2400 case (380W power supply with 85% tested efficiency), ECS Micro ATX motherboard, 2.0GHz Turion CPU, 2GB DDR RAM, 80GB 2.5" SATA HD, NVidia 7600 fanless video card. I also replaced all fans (92mm CPU and 120mm case) with Nexus low noise fans.
It would seem that, with the possible exception of the video card (which is near the bottom of the list of card supported by MythTV), I had the same ideas you do. It meets my own requirements of being a low noise computer, however it's not especially low power. This machine idles at 75W and uses 140W when playing a DVD in MythTV. That's pretty good for a modern computer, but nowhere close to a laptop. Maybe all of the excess over 20W/34W is the video card. Seems like a lot of power for a low-end card.
That's an easy way to get energy efficiency, but you can build a desktop more efficiently. 80PLUS PSU, Turion CPU, etc., and your desktop can be lower power than older notebooks, while significantly faster.
How old of a laptop are you talking about? My svn/trac/ftp/web/email server is a 2001 IBM Thinkpad A21p (850MHz P-M, 512MB RAM, 100GB 7200rpm drive) that draws 20W idling and 34W going full tilt (with the screen off, natch). According to the numbers on the power brick, my 1999 Thinkpad 600X had similar power consumption. As do the two 2005 Thinkpad R51p's (1.7GHz Centrino) I'm using as workstations.
There's no good to mediocre desktop PC (aside from the Mac mini, which is essentially a laptop in a different box) that consumes that little power with any kind of performance.
So, what kind of power consumption can you get from a desktop? If you can beat 34W while compiling the kernel and get it done quickly, you should stop whatever else you're doing and start a company assembling PC's. Please let me know if you do, I would be very interested to see what you come up with.
Actually, I've run into a lot of people who have problems with evolution even though they aren't Christian or religious.
I find this statement hysterical. 150 years ago, when Darwin was making waves, evolution wasn't in doubt. Nobody disputed the evidence for evolution: you can go look at the bodies of different but similar species from the past and the farther you go back, the more different they get. So things must have changed, i.e. evolution. What Darwin got everyone all excited about is the theory that natural selection (without the hand of God) was enough for evolution to work and to produce the modern richness of species visible on the planet.
In the 150 years since then, this debate has not advanced. It's regressed. These days in the US (and pretty much only in the US), 48% of the population not only doubts natural selection, but evolution itself. As in, you're choosing to doubt facts and conclusions that Darwin's critics had long accepted and put to rest while people were still using horses and carriages for transport and oil lamps for lighting.
Evolution is, to many, extremely unintuitive
Your statement, sir, is absolute nonsense. Only neo-Christian fundamentalists doubt evolution or that the earth is about 4 billion years old (give or take 500 million years). Well, them and some pacific islanders who have never seen someone with shite skin. Everyone else is lying (they're really neo-Christians but claiming otherwise).
my primary concern are the victims of drug abusers.
We already have to deal with intoxicated people operating cars, planes, and other potentially lethal machinery. How much worse would things be if now, in addition to those, you've got people high on ecstasy or marijuana?
By your logic (which I'm sympathetic to), the number of victims of users and addicts should roughly correlate with the number of users and addicts. So the goal of our public policy, including our laws, should be to reduce the number of users and addicts. Right? So, the first question that we should be able to answer is: has the presence of laws prohibiting marijuana, cocaine, heroin, methamphetamines, MDMA, etc. reduced the usage or addiction rate of any of those substances in the time since they were passed? You won't take my word for it, so I'll ask you to look it up for youself. The answer will probably suprise you.
What about heroin? Would bystander deaths double? Triple?
Or would they decline? The legalization of these drugs would reduce the price dramatically. After devastating the profitability of the black market, users can use cocaine and heroin in ways that are much less likely to cause addiction (taken orally instead of injected or heated and inhaled, for instance).
Last time I checked, not very many people grow tobacco in their backyards and make cigarettes in their basements. Why does anyone think dealers give the government a cut of their lucrative business?
There are at least two important differences between marijuana and tobacco. The profits on tobacco are much, much lower than the profits on marijuana. Also, marijuana grows like a weed in just about any medium and fixes nitrogen into it's roots (makes the soil richer), while tobacco is picky about soil conditions, and very damaging to the soil where it grows.
Society already has to pay for addicts, how many would we be paying for if these substances become easily and legally available?
And here's the core issue. I believe that just like making alcohol illegal increased drinking (and drunks), the current laws are largely responsible for the increased rates of users and addicts; and just like ending prohibition returned the number of drinkers and drunks down to pre-prohibition levels, legalizing drugs will reduce both users and addicts from their current numbers.
You seem to believe that alcohol usage rates around US alcohol prohibition, the changes in hard drug usage rates as hard drug prohibitions have gotten more and more severe, and the reduction in users and addicts in the Netherlands as they eliminate more and more prohibitions are the exceptions, and that all we need is more of what hasn't worked to finally fix the problem.
You and I both want fewer drug users and fewer drug addicts. The difference is that I'm willing to acknowledge that drug laws don't help and probably make that goal even more difficult. So I ask you, have we seen a substantial benefit or has the War on (Some) Drugs made the drug problem in this country worse?
Logic has no place in pro-drug arguments, because there is nothing logical about (ab)using these drugs in the first place.
There's a sneaky argument in there that you're not voicing. I'm not a user of any illegal drugs. Aside from the occasional glass of port, cup of tea, or Advil now and then, I don't use any drugs at all. I think that using addictive drugs is one of the stupidest possible things a person could do. And yet I firmly believe that legalization is the only chance we have to (1) reduce the number of drug users and addicts; (2) reduce the number of secondary crimes related to drug dealing and drug buying; (3) reduce the funding for gangs and other black-market organizations; and (4) begin the process of restoring some of our long-lost freedoms. We've paid for the War on (Some) Drugs with the fourth, fifth, ninth, and tenth amendments to the Constitution.
None, really, go against the basics of not murdering, or not stealing, or not lieing.
Lying is routinely moral, especially lying to prevent hurt feelings. Killing is not murder, which is really an astonishingly precise action, and killing as a soldier or killing in self-defense are normally considered moral. Stealing to feed yourself or others, especially as a response to someone else's legal theft (Robin Hood) is usually considered moral.
And before any arguments to the contrary -- war, politics, etc. -- each and every culture, historically and presently, forbade such things.
What you're referring to are legal/illegal things. Laws only occasionally interact with morality, and what is legal has almost nothing to do with what is moral. Confusing legality and morality is a common but fatal error. Fatal to your argument, that is.
NiMH batteries are almost entirely recycleable. The high-nickel product from currently primitive recycling processes can be used as an input to stainless steel manufacturing, but is only marginally economical for making new batteries. When used in making stainless steel, the recycled product displaces nickel that would be produced by new extraction, and IMHO, should be considered as a 100% offset to the environmental cost of the nickel used in the original construction of the battery.
Nickel is already expensive enough that if nickel-based battery production ramps up, the economic value of the nickel will make battery remanufacturing fully cost-effective. Additional research on the manufacturing and recycling processes are also likely to provide substantial cost improvements from where we stand today. I wouldn't be suprised to find in 5-10 years that nickel-based batteries enjoy the same "near-100%" closed-loop recycling ability that lead-acid batteries currently enjoy.
OpenWRT wasn't very practical. It only worked on really old hardware that wasn't in stores anymore.
I've only flashed two Linksys routers with DD-WRT, but my experience couldn't be more different from yours... One is three years old, but the other is a six month old WRT54GL, and both still work like a charm.
Even then, you needed exactly the right serial number revision. The serial numbers that worked were made in small quantities and virtually impossible to find.
On this point, you're simply misinformed. Almost all Version 2-4 Linksys WRT54G and all WRT54GL routers work with DD-WRT just fine. And even then, DD-WRT can be run on an enormous number of broadcom-based broadband routers. Linksys just happened to be the first to sell them really cheap.
Flashed a Linksys access point and bricked it. There was no JTAG or bootloader on the router to recover it.
This is a real risk, though if you follow the flashing procedures exactly the risk is pretty low (in fact, you're the first non DD-WRT developer I've heard of who has bricked one). Luckily, the best router for DD-WRT is only $45 shipped, so even if things do go wrong, you're not out a whole lot of money.
What's really needed is wireless router for desktop computers instead of attempts to reverse engineer Linksys routers just for the sake of being embedded.
Have you compared the power consumption of a small desktop running linux vs a broadband router? 150W vs. 10W is what my "kill-a-watt" says. 140W running 24/7 costs me about $15/month (southern California), so the Buffalo router pays for itself in reduced power bills in three months. Another perspective: over the course of a year I would save almost $180 in power bills. I don't know about you, but I can do a lot with $180 and I'd rather keep that money instead of handing it over to the power company.
I posted this comment to his blog entry protesting fair use:
If you want to write a critique for a book, it's often necessary to refer to specific portions of the book, by description and repetition of the actual content, so that someone wanting to read your critique doesn't have to go and buy the work themselves just to understand what you're arguing. Fair use protects the ability of everyone to do that, thus protecting the ability to discuss, criticise, parody, promote, etc.
Ultimately, the ability to publically discuss is the core of culture, and that's infinitely more valuable than any supposed interests of the original copyright holder. Being free from criticism is only bliss to the fool.
Though I doubt that it will happen any time soon, what you should learn from this whole affair is that you (like all of us) don't have control over anything, and further, you won't ever have control. The internet has indeed put a serious dent in copyright by reducing the marginal cost of copying data to $0.00. Nothing is going to "fix" that fact. All that will happen from here is much gnashing and yelling as various industries adjust to that new fact of reality.
BTW, I understand completely if you choose not to post this.
Sure, you can remove a government with enough force-of-arms, but how are you going to go about setting up a new government that is better than the old one?
Well, our government here in the US has a power leak (it's like a memory leak, it keeps asking for more and more power), so we just need to hit the reset button every few hundred years. We're getting to the point where a large group of people need to stand up and say that the US Constitution says what it says. Which would be better than what we have now.
Hopefully, you guys in the UK can do something similar.
This is a government that represents a significant percentage of the population (26%, I think someone here said). That means 26% of the population is opposed to a revolution, and you'll have to supress them or kill them. Not generally considered a good way of governing a country, now, is it?
You're overestimating the number who will take up arms to defend the government. It's not the same number who voted for them. During the US Revolutionary War, only about 33% of the population favored independence, and not only did they get their way, but they made a better government afterwards (better for the colonists, anyway). So there's at least one example with much worse odds than you're talking about.
It. Won't. Happen. Ever.
You may very well be right. Honestly, I hope for a bloodless revolution that somehow pushes the endless power grab back a few generations. Do I believe it will happen? Probably not.
In my experience, when building a significant database/Java app, you've got two choices:
Table data driven.
Domain model driven.
If you choose table data driven, you rely on one or two skilled database developers (it's a huge mistake to assume that even great DBA's have any clue at database design, though all will claim they do) to create a clean, performant data model. The Java portion of the system is basically just a procedural presentation layer for the database.
Pros:
You can make a very clean database, designed exactly the way your DBA's think it should be designed.
If your dev cycle takes more than a week to produce and deploy a patch to the app server code, it's fairly simple to change stored procedures, tables, and anything else in the database to respond to scalability issues in production without disrupting the running application.
This architecture usually offers more "knobs" with which to tune performance, and can usually be made to perform faster than the object driven system. (in my experience, YMMV...)
Cons:
Your Java devs will chafe because from a Java developer's perspective, the Java part of the app has a crap design. You're going to get constant whining about the lack of OO, interfaces, etc.
Your DBA's will have lots of opportunities to "fiddle", which can be very risky because although they can tune the living shit out of what the system is doing right now, but won't understand what their fiddling will do to the maintainability of the presentation (Java) layer.
Locating business logic sanely is an enormous challenge. This app is largely owned by your DBA's, but a significant fraction of the business logic should be in the Java because it will be easier to answer many questions there. Remember the DBA's fiddling and what that does to system maintainability?
It's hard to write good procedural code using an OO language. Skilled Java developers will perpetually be trying to write OO code, which will conflict with the role of the Java application: to be a procedural presentation layer.
The Java guys will eventually force some limitations on the DBA's, who will chafe that the Java guys don't understand data.
If you choose a domain model driven architecture, you rely on one or two people with both Java and database design skills (it's a huge mistake to assume that even great Java devs have the slightest clue about data modeling, you need to find people with both skills) to create a clean, performant data model along with a set of domain objects that represent that data in the Java portion of your app.
Pros:
Your Java can be Object Oriented, well-designed, interface/implementation separation, etc.
Your database can still be cleanly designed, though it won't look exactly the same as if DBA's had designed it. Basically, some normalization decisions will end up different because of the additional information available to the designer(s).
The Java group is happy.
The domain objects are loosely coupled to the tables/views instead of tightly coupled to the stored procedure result sets. This is good because changes to the system have to take both Java and RDBMS portions into account and the system likely to remain maintainable.
Cons:
No matter how much expertise your Java/database guys have, your DBA's will chafe, because:
they didn't design the database and they wouldn't have come up with the same design.
They don't have full freedom to use all of their DBA skills on the app tables.
Caveat: your best DBA's will realize that this means they can focus on higher-level database problems and will actually be happier.
The domain objects are loosely coupled to the tables/views instead of tightly coupled to the stored procedure result sets. This is bad because the clean break under the stored procedures allowed easier optimization of the database, separate from the app server.
AOL/Time-Warner enormously relies on linux and Samba all over the place. This may or may not help your case depending on what your boss thinks of AOL as a company...
Slashdot Mirror
Regards,
Ross
Personally, I think that most "wine experts" are overblown windbags who engage in the worst overuse of metaphors in modern language. However, I do happen to agree with them now and again: every single time I've had anything from Stag's Leap, it's been incredible. Wines, like many other issues of taste, are difficult to come to any lasting consensus. You would do well to continue to trust your own taste over anyone else's.
(If you're in the mood for advice, you would do yourself a favor to lose the bit about California wines not being world class. You just come across as uninformed.)
Regards,
Ross
I have enjoyed chocolate around the world, including Belgium and Switzerland and IMHO, craft made chocolates from the US (I'm specifically thinking of Sharffen Berger (Hershey), Ghirardelli (Lindt), and Dove (Mars)) are the equal of any chocolatier anywhere in the world.
Also, Cadbury and Kraft are in the process of ruining many European chocolate brands, so it's not like the rest of the world is free from crap chocolate.
Of course, my palate may not be as refined as yours, so I'm not expecting universal agreement on such a personal subject as chocolate. But there is excellent US-made chocolate. It is simply necessary to be somewhat discriminating.
Ross
So, I'm a wierdo. But I did manage to find a wife who agrees with me on noise, so I'm not alone, just outnumbered.
Less glibly: I would love to be able to eliminate my motorcycle tailpipe and make it completely silent. I've heard that this would make me less safe, but I've noticed that when driving, I've never heard a motorcycle coming up behind me. Even the ones with loud pipes.
Regards,
Ross
I'll admit it, I've done that very thing (copy some code in order to bootstrap a project). But I put all sorts of comments around tainted code, and I make damned sure that every single line of tainted code is rewritten before that code, or a product based on it, is released into the wild.
Theo and Marcus both come across as graceless and petulant children. At least Marcus decided to be childish quietly. Theo's ongoing rants about "the inhumanity" of it all just get hysterical after a few posts. Yes, the original email probably should have been more private. But the response from Theo is completely and utterly over the top.
Regards,
Ross
Collection of random thoughts in response to your post:
You're not thinking about the costs on an apples to apples basis. If you've decided to rebuild a car, how much does a new (remanufactured) drivetrain cost? Having just finished a frame-off Jeep Scrambler buildup with: a rebuilt chevy 350 engine through a new centerforce clutch into a rebuilt NV4500 5-speed and then into a new Atlas transfer case... all-up cost of the powertrain is about $12k. Pretty comparable, given that one is for extreme performance/durability and the other is for extreme efficiency.
What you're really paying for there is an EV drivetrain that doesn't throw away batteries. The solectria controller should give you very good life on your batteries. Chewing through batteries is a common medium term problem with EV converisons, and I have no intention of spending that much money over and over again.
You definitely start from a salvage car to minimize the cost of the whole project. There are cheap 914's to be had on eBay (mostly partially completed project cars). The original engines in those things sucked.
In the EV, I'm taking the time to refurbish/replace/rebuild as much of the drivetrain as I can to guarantee that the car will last as long as a "new" car, and therefore be worth the $16k total cost of the project. BTW, I'm using the "AC heavy manual transmission" kit from them, which is also $10k.
Regards,
Ross
The big issue with my commute is that my home and my office are on opposite sides of UCLA. UCLA has a huge commuter student population and is usually surrounded with nasty traffic on roads with 35mph signs that everyone ignores, choosing instead to go 50mph when they can. These roads also lack any sort of bicycle lane and usually have no parking lane or shoulder. If I stay away from UCLA I can travel on slower/safer roads, but going that far around UCLA means a much longer trip.
With the motorcycle, I can slide through traffic and then go as fast or faster as the other cars when traffic speeds up. The bicycle is even better at slipping through traffic, but when things speed up on these roads, it's pretty dangerous if you can't keep up.
Regards,
Ross
20 x $160 is the pack for this car. Less the core charge, it comes to right around $3000. And given local gas prices (>$3/gal), I suspect that compact cars cost more per mile around here.Possible, and I looked at that, but the roads that I have to traverse have typical speeds of 45-50mph, and there are no alternate/slower routes that don't also more than double the distance.
At the moment, I ride a motorcycle and spend $12 every two weeks on gas to travel ~160 miles, which comes to about $0.08/mile. I agree that the EV will cost more than the motorcycle per mile, and that's been a question about this project from the beginning. The other issue is that I like riding the motorcycle and getting to work that much faster (lane splitting is legal in CA, so I spend 15 minutes instead of the 25-35 minutes it take on those days when I take the car for whatever reason).
Being really honest, at least half of the fun is the project itself.
Regards,
Ross
The 85 mile range is stated as conservative based on the very similar cars I'm using as sources for my design. One is a '93 Saturn SL2 (almost exactly the same curb weight), and when he drives carefully, he can get 90 miles before the controller begins complaining, giving him five more miles to get home with his setup. I'm using a higher efficiency motor than him, but also one fewer 12V batteries, so 85-90 miles per charge doesn't seem particularly ambitious.
Regards,
Ross
Not to say what you're talking about will never happen (though I'm extremely skeptical), but for 95%+ of my driving, plugging in overnight works great. Actually, since my commute is only 7 miles each way, I'll only need to charge for about two hours to get back to 80%+. Now, if I need to drop the kids off at a couple of activities, get groceries, do a few other errands, and take the family out for an evening's entertainment? The car will need to charge the whole time that I'm asleep and be topped off with a full charge when I wake up in the morning.
Electric cars are a lot more practical than most people think. Mine will cost me $15k, take 200 hours of my time, will do 80mph and get 85 miles on a charge that costs me $3.50 while hauling four people or light hauling. Some might want to include the 200 hours in the cost, however, I won't bill myself for the time since I find it so enjoyable to work on it and would pay extra to have this much fun
Regards,
Ross
The original plan for the train had three stops inside the U-shaped terminal, and there's even a sub-level of the LAX buildings that was originally designed to accomodate the eventual train system. So what happened? The taxi drivers said that if the train was in the terminal, their business would be seriously hurt. So the train stops two miles from the terminal, and you get to stand outside and wait for a bus to take you the rest of the way. There are simply too many political interests with the clout to make sure that public officials don't threaten their business, even if they are part of the real problem.
I don't travel that often, but when I do, it's a rather big PITA to figure the best way to get to the airport. Guess why I never take the train. Because the train sucks, at the request of the taxi drivers? You got it in one! Yeah, I'm only a little bitter about that.
Regards,
Ross
I'd say to try again with the numbers, perhaps with a vehicle more suitable for EV conversion. Starting from a compact car, EV's are MUCH more practical than what you're talking about.
Regards,
Ross
If you know of more information on the web similar to http://www.spcr.com/ that would help me learn more on this subject, please post it. I'm definitely interested!
Regards,
Ross
It would seem that, with the possible exception of the video card (which is near the bottom of the list of card supported by MythTV), I had the same ideas you do. It meets my own requirements of being a low noise computer, however it's not especially low power. This machine idles at 75W and uses 140W when playing a DVD in MythTV. That's pretty good for a modern computer, but nowhere close to a laptop. Maybe all of the excess over 20W/34W is the video card. Seems like a lot of power for a low-end card.
Regards,
Ross
There's no good to mediocre desktop PC (aside from the Mac mini, which is essentially a laptop in a different box) that consumes that little power with any kind of performance.
So, what kind of power consumption can you get from a desktop? If you can beat 34W while compiling the kernel and get it done quickly, you should stop whatever else you're doing and start a company assembling PC's. Please let me know if you do, I would be very interested to see what you come up with.
Regards,
Ross
In the 150 years since then, this debate has not advanced. It's regressed. These days in the US (and pretty much only in the US), 48% of the population not only doubts natural selection, but evolution itself. As in, you're choosing to doubt facts and conclusions that Darwin's critics had long accepted and put to rest while people were still using horses and carriages for transport and oil lamps for lighting.Your statement, sir, is absolute nonsense. Only neo-Christian fundamentalists doubt evolution or that the earth is about 4 billion years old (give or take 500 million years). Well, them and some pacific islanders who have never seen someone with shite skin. Everyone else is lying (they're really neo-Christians but claiming otherwise).
Regards,
Ross
You seem to believe that alcohol usage rates around US alcohol prohibition, the changes in hard drug usage rates as hard drug prohibitions have gotten more and more severe, and the reduction in users and addicts in the Netherlands as they eliminate more and more prohibitions are the exceptions, and that all we need is more of what hasn't worked to finally fix the problem.
You and I both want fewer drug users and fewer drug addicts. The difference is that I'm willing to acknowledge that drug laws don't help and probably make that goal even more difficult. So I ask you, have we seen a substantial benefit or has the War on (Some) Drugs made the drug problem in this country worse?There's a sneaky argument in there that you're not voicing. I'm not a user of any illegal drugs. Aside from the occasional glass of port, cup of tea, or Advil now and then, I don't use any drugs at all. I think that using addictive drugs is one of the stupidest possible things a person could do. And yet I firmly believe that legalization is the only chance we have to (1) reduce the number of drug users and addicts; (2) reduce the number of secondary crimes related to drug dealing and drug buying; (3) reduce the funding for gangs and other black-market organizations; and (4) begin the process of restoring some of our long-lost freedoms. We've paid for the War on (Some) Drugs with the fourth, fifth, ninth, and tenth amendments to the Constitution.
Regards,
Ross
Regards,
Ross
NiMH batteries are almost entirely recycleable. The high-nickel product from currently primitive recycling processes can be used as an input to stainless steel manufacturing, but is only marginally economical for making new batteries. When used in making stainless steel, the recycled product displaces nickel that would be produced by new extraction, and IMHO, should be considered as a 100% offset to the environmental cost of the nickel used in the original construction of the battery.
Nickel is already expensive enough that if nickel-based battery production ramps up, the economic value of the nickel will make battery remanufacturing fully cost-effective. Additional research on the manufacturing and recycling processes are also likely to provide substantial cost improvements from where we stand today. I wouldn't be suprised to find in 5-10 years that nickel-based batteries enjoy the same "near-100%" closed-loop recycling ability that lead-acid batteries currently enjoy.
Regards,
Ross
Regards,
Ross
I posted this comment to his blog entry protesting fair use:
If you want to write a critique for a book, it's often necessary to refer to specific portions of the book, by description and repetition of the actual content, so that someone wanting to read your critique doesn't have to go and buy the work themselves just to understand what you're arguing. Fair use protects the ability of everyone to do that, thus protecting the ability to discuss, criticise, parody, promote, etc.
Ultimately, the ability to publically discuss is the core of culture, and that's infinitely more valuable than any supposed interests of the original copyright holder. Being free from criticism is only bliss to the fool.
Though I doubt that it will happen any time soon, what you should learn from this whole affair is that you (like all of us) don't have control over anything, and further, you won't ever have control. The internet has indeed put a serious dent in copyright by reducing the marginal cost of copying data to $0.00. Nothing is going to "fix" that fact. All that will happen from here is much gnashing and yelling as various industries adjust to that new fact of reality.
BTW, I understand completely if you choose not to post this.
Regards,
Ross
Hopefully, you guys in the UK can do something similar.You're overestimating the number who will take up arms to defend the government. It's not the same number who voted for them. During the US Revolutionary War, only about 33% of the population favored independence, and not only did they get their way, but they made a better government afterwards (better for the colonists, anyway). So there's at least one example with much worse odds than you're talking about.You may very well be right. Honestly, I hope for a bloodless revolution that somehow pushes the endless power grab back a few generations. Do I believe it will happen? Probably not.
Sad, really.
Ross
If you choose table data driven, you rely on one or two skilled database developers (it's a huge mistake to assume that even great DBA's have any clue at database design, though all will claim they do) to create a clean, performant data model. The Java portion of the system is basically just a procedural presentation layer for the database.
Pros:
Cons:
If you choose a domain model driven architecture, you rely on one or two people with both Java and database design skills (it's a huge mistake to assume that even great Java devs have the slightest clue about data modeling, you need to find people with both skills) to create a clean, performant data model along with a set of domain objects that represent that data in the Java portion of your app.
Pros:
Cons:
AOL/Time-Warner enormously relies on linux and Samba all over the place. This may or may not help your case depending on what your boss thinks of AOL as a company...
Ross