That's it I'm off. You've driven me out of the house into the pissing rain where I'm going to get soaking wet... I hope you're happy with yourselves. I'll have to mingle with... "people"... I may catch something, if I do, I'll blame you.
No, my analogy is accurate. Not when you're comparing a large capital cost against a small service cost. It makes absolutely no sense to invest $100k in a power system to avoid a $1,500 annual expense... Which is essentially what this project does.
Or weren't you being ironic, I assumed you were because looking at your analogy with respect to the article, yes, the finished product "drywall" is a dumb idea.
God the problems I had trying to handle large datasets... Where "large" is bigger than say 64k... So what I really mean by large is small. Excel is just completely useless for anything non trivial.
Yes as you mentioned, there are better tools for the job and frankly as hard as they might seem, they just work.
Parent is very wrong. I'm one of a couple of devs in my office using Ubuntu as my desktop. I use Open Office and can open all docs that people send to me: Powerpoint, Excel, Word docs And thought "Oh FUCK! Time to bring out a new "improved"(har har) and incompatible version of Office.". So there you have it.
Your Open Office system will work fine for about 18 months until the new version starts to become more common, then you (and every other existing MS Office user as well) will start running into problems as the network effect with the new version really kicks in.
This isn't anything new per se, just that the complexity of the modelled systems is getting larger, and due to the numercal estimation processes needed to get anything remotely usable these realms haven't been accessible until lately with the increase of computing power http://boinc.berkeley.edu/
Suppose you had this revolutionary idea called "drywall". To make your analogy more accurate...
Suppose you work out that after development costs are accounted for, it still costs $100,000 to "drywall" a house and only $5,000 to plaster it normally.
Lets see... Desktop PCs, servers, Office Suite, Accounts package, RDBMS, web servers, application server, security software, Windows administrator, server administrator, database administrator... It all adds up.
Now imagine Google does it all for you, all you need is a client application and a network connection. They guarantee performance, security and that your data is yours when you want it. All for $20 per month per seat.
Today you pay that rate. Don't assume that energy costs won't continue to rise over the next 10-20 years. And at some point it may be worth going green. It just isn't at the moment.
If the world power consumption stays on its current growth rate, and if anything it's poised to accelerate, then by 2040 we will need to double the worlds energy production. Um, nope... Our use of energy, and roads increase to fill existing capacity. You could double the amount of energy produced tomorrow, and the number of roads and what'd happen is that our use would simply double to fill it. If we cap our energy production or roads, our use will remain static at the current level. If you reduce the energy production or roads, we will simply use them more efficiently.
The key concept is that it's a general principle. We tend towards the use of all available capacity of a resource. Be it oil, water, energy, roads whatever. It's why btw, we won't end up with perpetual gridlock in the future, no matter what we do to our roads.
My point was that people don't go green because it costs more to do so. If you want them to go green you have to make it cheaper for them than the status quo. The environmental problems we have are not technical, they never have been, they're economic. They can only be solved through economic measures.
And that, in a nutshell, is why creationism can be believed by otherwise intelligent people. It is? Evolution is a numbers game.
Evolution is slow, so slow that we can't really conceive of how slow that is. Slow? Is it now?
Evolution proceeds at different rates depending on the environment. If a population of long, medium and short tailed monkeys has all of the long tailed ones killed in a generation, then the species has evolved in that one generation... 5 years? The genes for long tails have gone.
Random mutations now... Well that depends on just how good the copying process is, but no matter how good it is, it isn't perfect. There's no such thing as perfection in the real world so errors are going to appear and accumulate in offspring.
As to those probes. How does it take for a million self-replicating probes to become a trillion. Depends how many offspring each generation have on average, the time between generations and available resources.... And if mutated machines predate on one another... No reason to believe they wouldn't.
But if someone's going to make predictions about the next billion years you can guarantee that the predictions will be wrong in ways that we can't even begin to imagine. Sure, but look at it this way. Where are all the self replicating machines?
but who develops the sophisticated software that run super computer simulations? The CS major. Does it need lots of people who develop super computer simulations?
How's about a modular approach? Let students choose what they think they need.
Can you imagine humanity restrained enough that over the next few million years of history, with a population of trillions spread over many planets with a multitude if different cultures I think you mean species. But no, I don't and I fully expect that in the medium future, some species of humanity will end up in a war with what're essentially machines.
The question is moot anyway - the distinction between machine and organism will become pretty blurred over the next few millennia. Mmm. in the very long term machines will be replaced by genetic modifications or biological organisms. Machines require too much energy; an organic (carbon based) being can always outcompete a metal/silicon being because it requires less energy. With the short term oil boom we're in, energy doesn't matter so much but in the very long term, it very much does.
Or, you could just program them to destroy themselves if the has of their program doesn't match. That would only work until that particular part of the system was corrupted during replication process. Evolution is inevitable as soon as you have analogue copying. It might take a million replications for that particular error to hit, but it'd happen eventually.
He rejects self-replicating probes because they'd compete with the original explorers. I think that's a lame argument What makes it lame? We're talking physical, analogue copies here. That means cumulative errors in every succeeding generation of machines, this means the machines will evolve. They will then compete for resources with anything else which occupies the same space. The same's true of human colonies btw... Look at America competing for oil resources with the EU.
Self replicating is ruled out due to risk. That sounds fairly silly since computers are computers. They do what we tell them to and not a thing more. Every copy of something introduces error. The instant you introduce a physical replicating system in the real world you also introduce evolution. At which point it stops working for you and starts working for itself.
Von Neumann machines will colonize the galaxy in direct competition with the species creating them, they won't explore for them.
Slashdot Mirror
I simply don't get any.
That's it I'm off. You've driven me out of the house into the pissing rain where I'm going to get soaking wet... I hope you're happy with yourselves. I'll have to mingle with... "people"... I may catch something, if I do, I'll blame you.
Or weren't you being ironic, I assumed you were because looking at your analogy with respect to the article, yes, the finished product "drywall" is a dumb idea.
With an axe... Obviously.
God the problems I had trying to handle large datasets... Where "large" is bigger than say 64k... So what I really mean by large is small. Excel is just completely useless for anything non trivial.
Yes as you mentioned, there are better tools for the job and frankly as hard as they might seem, they just work.
Which is very very handy indeed.
Your Open Office system will work fine for about 18 months until the new version starts to become more common, then you (and every other existing MS Office user as well) will start running into problems as the network effect with the new version really kicks in.
Suppose you work out that after development costs are accounted for, it still costs $100,000 to "drywall" a house and only $5,000 to plaster it normally.
Then yes "drywall" would be a bloody stupid idea.
Lets see... Desktop PCs, servers, Office Suite, Accounts package, RDBMS, web servers, application server, security software, Windows administrator, server administrator, database administrator... It all adds up.
Now imagine Google does it all for you, all you need is a client application and a network connection. They guarantee performance, security and that your data is yours when you want it. All for $20 per month per seat.
UYFB: What type of energy does a fuel cell produce?
Ahh cause and effect, supply and demand. How misunderstood they are...
The key concept is that it's a general principle. We tend towards the use of all available capacity of a resource. Be it oil, water, energy, roads whatever. It's why btw, we won't end up with perpetual gridlock in the future, no matter what we do to our roads.
And drive an electric car?
. stm
What? No? I guess you'll be happy to have him over in Iraq raping and murdering minors then.
http://news.bbc.co.uk/1/hi/world/americas/6277145
They're guessing at the $100,000 figure btw.
My point was that people don't go green because it costs more to do so. If you want them to go green you have to make it cheaper for them than the status quo. The environmental problems we have are not technical, they never have been, they're economic. They can only be solved through economic measures.
http://www.whynot.net/ideas/2195
Haranguing people to be green is useless... and annoying. Christ, they've been at it for 20 years. It's simple. Give people a worthwhile incentive.
Hmm?
And this is the reason so few people (including me) are "green".
http://en.wikipedia.org/wiki/Peppered_moth_evolut
http://www.dinosauria.com/jdp/evol/lizard.html
http://www.nytimes.com/2006/03/07/science/07evolv
Evolution proceeds at different rates depending on the environment. If a population of long, medium and short tailed monkeys has all of the long tailed ones killed in a generation, then the species has evolved in that one generation... 5 years? The genes for long tails have gone.
Random mutations now... Well that depends on just how good the copying process is, but no matter how good it is, it isn't perfect. There's no such thing as perfection in the real world so errors are going to appear and accumulate in offspring. As to those probes. How does it take for a million self-replicating probes to become a trillion. Depends how many offspring each generation have on average, the time between generations and available resources.... And if mutated machines predate on one another... No reason to believe they wouldn't.
The CS major. Does it need lots of people who develop super computer simulations?
How's about a modular approach? Let students choose what they think they need.
"Yes, you're allowed to get drunk off your ass and purchase hardcore porno flicks, but you can't buy these video games"
Why do people continue to conflate sex and violence? One is about having fun and procreating, the other is about hurting people.
Von Neumann machines will colonize the galaxy in direct competition with the species creating them, they won't explore for them.