If you RTFA, it is the driver who plowed into the pedestrians. He assumed the car would detect the pedestrians and slam on the brakes for him, but it turns out the car did not have the feature or it was disabled. A stupid, dangerous stunt on the part of the people involved, and some seriously dishonest reporting.
If you RTFA, its pretty obvious Anaya knew he was helping criminals. I don't feel sorry for him at all. If he really wanted to keep a clean business he could have said that he would document all installations and share them with the cops.
There's nothing wrong with this picture. The Arduino is for hobbyists to learn and play, not for designing cost effective production grade products. If Arduino users can do DSP on their boards, the more power to them.
Note that the upcoming Arduino Due is a Cortex-M3 (32 bit, 96MHz, no-FPU); this thing could do a fair amount of DSP compared to older Arduinos.
It's easier for people to approach problem solution in a procedural way than it is for them to think about it functionally. And that's why functional languages, no matter how elegant or "great" they may be, will never really break into the mainstream.
What's easiest is what you're used to. The most popular programming system on earth is Excel, which is firmly in the functional camp.
Or just pump it out of the barge over the period of a month or so while driving around the Pacific. Seawater is already naturally radioactive, and I doubt spreading this barge-full of radioactive water around would make any difference.
One problem for ARM is that every ARM processor and board has different hardware, and that hardware is in different places. Compare this to an x86 PC where you are guaranteed to have basic components at known locations; enough so that a generic OS image can boot. An ARM OS image typically needs a lot of static configuration, custom drivers, board specific bootloader, etc.
Does the GPL require the source code to run on the same piece of hardware? The OP can't build an run new firmware on his router, but can he build and run it on his x86 linux machine with standard tools? If *that* can be done, is it really still a violation? The modified source code has been re-contributed to society.
I know that's not really what the OP wants to accomplish though...
What those other FP languages have that Lisp doesn't is a clean concept of *pattern matching*. Sure, you can do pattern matching in Lisp (see Termite Scheme), but its much more verbose than languages that have the syntax and semantics for it built-in.
I agree that Haskell is hard to understand, but only because it is easy to make incredibly dense code. Reading one line of Haskell can load your brain as much as reading 12 lines of anything else....And monads are just plain hard.
All that said, as an intermediate Erlang programmer, I am often left yearning for the cleanliness of Haskell syntax and also for S-expressions when I get into metaprogramming.
Even if the block heater plugs can't put out a lot of juice, trickling power over 8 hours while you are at work will make a difference. But the owners of those plugs would need to recognize that they being used for charging, and would probably want to charge for power consumed. As the parent mentions, they are often turned off in the summer, and many places will only turn them on say 20 minutes out of every hour in the winter (deadbeats!).
Plug in hybrids would be great for Manitobans... we've got some of the cheapest rates in North America (5.94 cents/kwh, 5.69 in US cents), and almost all of it is hydroelectric.
This kind of product has been commercialized in select cities in Canada for a while now. The Otto has a map of speed limits, intersections, photoradar, etc.
http://www.myottomate.com/
The theory is that functional langauges are far more parallelizable than imparative langauges like C and java.
In pure functional programming you say what you need done, but the runtime system can do what needs to be done in any order it feels like, including chopping it into little pieces and feeding it to many CPUs.
I don't know how many of today's functional languages can be parallelized. Scheme and Ocaml have a lot of imperative features that would gum up the works. Languages like Haskell show more promise.
Personally, I think imperative languages will fall out of the mainstream in the coming years (decades?). C programs like Apache might work fine on 64 cores, but how well will it work on a million core cpu?
Math is forever, Von-Neuman style computers aren't.
I thought I read some time ago that Simon Peyton Jones, a key designer of the Haskell language and Microsoft employee, was involved with this new shell. Hence the name "Monad". Of course I can't find that reference now...
Slashdot Mirror
If you RTFA, it is the driver who plowed into the pedestrians. He assumed the car would detect the pedestrians and slam on the brakes for him, but it turns out the car did not have the feature or it was disabled. A stupid, dangerous stunt on the part of the people involved, and some seriously dishonest reporting.
On the other hand, Soyuz hasn't lost a crew in over 40 years. I'd call that a pretty good safety record.
Would go great with Kerbal Space Program.
SpaceShipTwo has a top speed of approx 1200 m/s, whereas low earth orbit is on the order of 8000 m/s.
If you RTFA, its pretty obvious Anaya knew he was helping criminals. I don't feel sorry for him at all. If he really wanted to keep a clean business he could have said that he would document all installations and share them with the cops.
Sounds analogous to the light gas gun.
One can do 400m in KSP. If you're not snapping off spaceship parts you're too far away.
- A 747 consumes 140MW. [ http://en.wikipedia.org/wiki/Orders_of_magnitude_(power) ]
- Nevada Solar One, a 400 acre solar generating station, generates 64MW. [ http://en.wikipedia.org/wiki/Nevada_Solar_One ]
Hmmmm...
Scratch, visual multimedia programming system from MIT. http://scratch.mit.edu/
DSP with Arduino
What is wrong with this picture?
There's nothing wrong with this picture. The Arduino is for hobbyists to learn and play, not for designing cost effective production grade products. If Arduino users can do DSP on their boards, the more power to them. Note that the upcoming Arduino Due is a Cortex-M3 (32 bit, 96MHz, no-FPU); this thing could do a fair amount of DSP compared to older Arduinos.
It's easier for people to approach problem solution in a procedural way than it is for them to think about it functionally. And that's why functional languages, no matter how elegant or "great" they may be, will never really break into the mainstream.
What's easiest is what you're used to. The most popular programming system on earth is Excel, which is firmly in the functional camp.
Or just pump it out of the barge over the period of a month or so while driving around the Pacific. Seawater is already naturally radioactive, and I doubt spreading this barge-full of radioactive water around would make any difference.
One problem for ARM is that every ARM processor and board has different hardware, and that hardware is in different places. Compare this to an x86 PC where you are guaranteed to have basic components at known locations; enough so that a generic OS image can boot. An ARM OS image typically needs a lot of static configuration, custom drivers, board specific bootloader, etc.
Does the GPL require the source code to run on the same piece of hardware? The OP can't build an run new firmware on his router, but can he build and run it on his x86 linux machine with standard tools? If *that* can be done, is it really still a violation? The modified source code has been re-contributed to society. I know that's not really what the OP wants to accomplish though...
Here's a tutorial to get you rolling with OpenGL in SDL under linux. Once you get rolling with that, you can proceed to the NeHe tutorials. http://cone3d.gamedev.net/cgi-bin/index.pl?page=tutorials/ogladv/tut1
What those other FP languages have that Lisp doesn't is a clean concept of *pattern matching*. Sure, you can do pattern matching in Lisp (see Termite Scheme), but its much more verbose than languages that have the syntax and semantics for it built-in.
...And monads are just plain hard.
I agree that Haskell is hard to understand, but only because it is easy to make incredibly dense code. Reading one line of Haskell can load your brain as much as reading 12 lines of anything else.
All that said, as an intermediate Erlang programmer, I am often left yearning for the cleanliness of Haskell syntax and also for S-expressions when I get into metaprogramming.
Even if the block heater plugs can't put out a lot of juice, trickling power over 8 hours while you are at work will make a difference. But the owners of those plugs would need to recognize that they being used for charging, and would probably want to charge for power consumed. As the parent mentions, they are often turned off in the summer, and many places will only turn them on say 20 minutes out of every hour in the winter (deadbeats!). Plug in hybrids would be great for Manitobans... we've got some of the cheapest rates in North America (5.94 cents/kwh, 5.69 in US cents), and almost all of it is hydroelectric.
This kind of product has been commercialized in select cities in Canada for a while now. The Otto has a map of speed limits, intersections, photoradar, etc. http://www.myottomate.com/
The theory is that functional langauges are far more parallelizable than imparative langauges like C and java. In pure functional programming you say what you need done, but the runtime system can do what needs to be done in any order it feels like, including chopping it into little pieces and feeding it to many CPUs. I don't know how many of today's functional languages can be parallelized. Scheme and Ocaml have a lot of imperative features that would gum up the works. Languages like Haskell show more promise. Personally, I think imperative languages will fall out of the mainstream in the coming years (decades?). C programs like Apache might work fine on 64 cores, but how well will it work on a million core cpu? Math is forever, Von-Neuman style computers aren't.
I thought I read some time ago that Simon Peyton Jones, a key designer of the Haskell language and Microsoft employee, was involved with this new shell. Hence the name "Monad". Of course I can't find that reference now...