Could be worse... I know for a fact that recent Boeings have their flight control systems programmed in a highly constrictive subset of Ada... although at least the programmers weren't forced to use vi.
There's a finite number of candidates coming out of MIT, CalTech, etc. Many of the undergrads from these institutions go straight to grad school - they're not part of the job market until they have a Master's or PhD.
All that said - look at the curriculum of each and see if it allows a broad span of theory and practice. You need to understand how things work, but you also need to spend some time slinging code and figuring out how to be personally productive and what areas of software development you like and enjoy most.
And when all that's done, be picky for grad school instead:)
THIS is why I read Slashdot... not for the keen, insightful commentary, or the devilish, bleeding-edge wit... no, it's for the community to geek out with.
Now, if we were to design such a mission... no, no, no. Just let it die. The Shuttle's the wrong thing to take out of LEO.
I watched the webcast. Never saw an Atlas V before - impressive. They kicked over to a "visualization" derived from the live telemetry after first-stage MECO - a good use of computing cycles.
Remember, this is an EXPERIMENT. It's not designed to roam around in orbit forever, servicing whatever satellites it meets, that look nice, that have "compatible adapters"... Orbital Express is a mission with 2 satellites, designed to interoperate, and the target (NextSat) basically sits there stable while the active vehicle (ASTRO) does all the hard work. They're not kicking off a business of servicing satellites - they're proving it can be done, and how hard it will be.
The business case for on-orbit refuelling I leave to simple Google searches.
Not to mention that in the months of traversing the Van Allen belts the astronauts, the shuttle avionics, and any thing else susceptible to radiation will get fried.
There's a darn good reason the Apollo missions blew through MEO quickly. The environment isn't very nice for humans between the lower Van Allen and GEO.
I'm not sure the poster really read all the article - it's not like this guy is the first to invent a gigapixel camera. Really, what he's build is a seemingly grain-free, "infinite focus" camera that takes large-format negatives, then he develops them (slowly, by hand) - THEN they get scanned at gigapixel resolution, and still show incredible detail.
You can look at his film prints "with your nose up to them", or you can scan them and zoom in on features down to the resolution of your scanner, apparently. Something innovative with basic cameras, film, and optics - not so innovative with the scanners and "gigapixel" resolution.
No company voluntarily "loses" a multi-million dollar functional asset. If it's still got functional transponders, they'll keep using it.
As for your other point - when possible, close to end-of-life, they try to move geosync birds to a super-synchronous (above the geosync plane) orbit, which will eventually cause them to migrate to nodal points safely out of the way of the remaining commercial satellites. This is often accomplished by a thruster burn that exhausts the remaining fuel in the tanks (preventing later tank explosions after thermal control is lost). If the satellite fails before planned end-of-life (usually determined by available fuel or power), it will end up in a figure-8 orbit roughly centered on the equator, and will slowly drift East or West depending on whether it was low or high, causing collision-avoidance issues for the rest of the geostationary com birds out there.
There's a lot of reference material out there - give it a read.
Fine. Mathemeticians can optimize an algorithm at a pace roughly analogous to the speed increase in hardware.
Which version of Windoze do you want to run it on?
There's Software written by people with Purpose, then there's the stuff that gets sold in shrink-wrap (or preinstalled), that makes a Pentium 4 run like a P2. That's not algorithm development - that's eye-candy and CPU-munching filesystem indexers.
Any language can "leak" memory, depending on how the programmers program. See this IBM developerWorks note for a how-it-happens, how-it's-cured tutorial. If Java didn't leak memory, how would JProfiler exist?
Few of us here are lawyers (I know I ain't), but aren't corporations considers "persons" for legal purposes? "I, IBM, take thee, Red Hat, to be my lawful flunky..."
or even worse: "I, Ben Affleck's Career, take thee, Jennifer Garner's Career, to be my legal co-producer, to have meetings with, schedule lunches, and arrange occasional carnal interludes with and between our respective Parties of the First Part...."
Gold has value because there is a standing market for it, in which people trade, and supply and demand have set a price. As the author of the paper states, gold is desirable because the supply is stable and the market is well-defined - it has industrial, cosmetic, and financial uses, all of which work to establish a price.
Platinum, he notes, is not seen as a "currency" because even though it's worth more per unit weight than gold (in the current market), the supply is unstable - new industrial uses for platinum are coming out all the time, and new sources and refining methods are constantly being worked. It's too unstable to be a worthwhile currency.
It's a "social Science" because they follow the scientific method as well as they can, but it's nigh unto impossible to actually perform meaningful experiments in economics above the micro- level (unless you're a world oil power, in which case you can experiment with demand-curves all you want). That's one of the weaknesses of the scientific method when applied to things above the biologcal scale - you can come up with the hypothesis, you can even come up with the experiment - but for it to be meaningful, you'd have to persuade a few thousand people to take part without modifying any other elements of their behavior. Tough for them.
This same guy was featured in one of the McGraw-Hill aviation mags (maybe Aviation Week?) preaching the same thing over 6 months ago - heck, I submitted it as a story then (Rejected, of course). He's watching his business base evaporate.
Green Hills can still make money if they debug their software, work on tools and add platforms - areas where F/OSS doesn't have the depth - esp. in niche processors.
The main cost driver in getting a PP-ASEL (I'm post-solo, so I know) is the cost of plane and instructor time. If this only requires 20 hours, versus 40 required for the Private Pilot, then that's a big stack of cash right there, esp. at $145/hr for dual time, $100/hr for solo in the C172-R I fly.
Wonder if my school is going to get any of these planes....
If the plane doesn't come from the manufacturer with specific equipment installed, the time and expense of the paperwork required to get approval to install new equipment, get the installation inspected and approved, and then get authorized to actually use said equipment in the air adds up quickly. This is non-trivial in the US (think of adding an IFR-capable GPS to an old piston twin to be used for charter service), but I understand it's nigh unto impossible in the EU right now.
Slashdot Mirror
Could be worse... I know for a fact that recent Boeings have their flight control systems programmed in a highly constrictive subset of Ada... although at least the programmers weren't forced to use vi.
There's a finite number of candidates coming out of MIT, CalTech, etc. Many of the undergrads from these institutions go straight to grad school - they're not part of the job market until they have a Master's or PhD.
:)
All that said - look at the curriculum of each and see if it allows a broad span of theory and practice. You need to understand how things work, but you also need to spend some time slinging code and figuring out how to be personally productive and what areas of software development you like and enjoy most.
And when all that's done, be picky for grad school instead
THIS is why I read Slashdot... not for the keen, insightful commentary, or the devilish, bleeding-edge wit... no, it's for the community to geek out with.
Now, if we were to design such a mission... no, no, no. Just let it die. The Shuttle's the wrong thing to take out of LEO.
Ares/Orion, now....
I watched the webcast. Never saw an Atlas V before - impressive. They kicked over to a "visualization" derived from the live telemetry after first-stage MECO - a good use of computing cycles.
Remember, this is an EXPERIMENT. It's not designed to roam around in orbit forever, servicing whatever satellites it meets, that look nice, that have "compatible adapters"... Orbital Express is a mission with 2 satellites, designed to interoperate, and the target (NextSat) basically sits there stable while the active vehicle (ASTRO) does all the hard work. They're not kicking off a business of servicing satellites - they're proving it can be done, and how hard it will be.
The business case for on-orbit refuelling I leave to simple Google searches.
Not to mention that in the months of traversing the Van Allen belts the astronauts, the shuttle avionics, and any thing else susceptible to radiation will get fried.
There's a darn good reason the Apollo missions blew through MEO quickly. The environment isn't very nice for humans between the lower Van Allen and GEO.
Um, yeah, dude, 'cause that middle word is "Barrada". Oh, crap, look what you've done now....
Remember - compilers don't care about comments!
I'm not sure the poster really read all the article - it's not like this guy is the first to invent a gigapixel camera. Really, what he's build is a seemingly grain-free, "infinite focus" camera that takes large-format negatives, then he develops them (slowly, by hand) - THEN they get scanned at gigapixel resolution, and still show incredible detail.
You can look at his film prints "with your nose up to them", or you can scan them and zoom in on features down to the resolution of your scanner, apparently. Something innovative with basic cameras, film, and optics - not so innovative with the scanners and "gigapixel" resolution.
Or maybe I read the whole NYTimes thing wrong....
No company voluntarily "loses" a multi-million dollar functional asset. If it's still got functional transponders, they'll keep using it.
As for your other point - when possible, close to end-of-life, they try to move geosync birds to a super-synchronous (above the geosync plane) orbit, which will eventually cause them to migrate to nodal points safely out of the way of the remaining commercial satellites. This is often accomplished by a thruster burn that exhausts the remaining fuel in the tanks (preventing later tank explosions after thermal control is lost). If the satellite fails before planned end-of-life (usually determined by available fuel or power), it will end up in a figure-8 orbit roughly centered on the equator, and will slowly drift East or West depending on whether it was low or high, causing collision-avoidance issues for the rest of the geostationary com birds out there.
There's a lot of reference material out there - give it a read.
"segway"? You're going to ride a Segway into a snapping towel, swinging a bullwip?
Fine. Mathemeticians can optimize an algorithm at a pace roughly analogous to the speed increase in hardware.
Which version of Windoze do you want to run it on?
There's Software written by people with Purpose, then there's the stuff that gets sold in shrink-wrap (or preinstalled), that makes a Pentium 4 run like a P2. That's not algorithm development - that's eye-candy and CPU-munching filesystem indexers.
Any language can "leak" memory, depending on how the programmers program. See this IBM developerWorks note for a how-it-happens, how-it's-cured tutorial. If Java didn't leak memory, how would JProfiler exist?
Link for http://www.real.com/linux/?src=092804realhome_1_1_ 2_1_1_1,021204r1cp_home_dlrhap_bb
RealPlayer 10 for Linux
OMG... did I just slashdot the webcast I was watching?
Bad Coz, bad Coz...
There's a webcast link from the people actually sponsoring it (who presumably know more than the normal press:
XPrize.Org
OB Princess Bride reference:
Inconceivable
"I do not think that means what you think it means."
Although it does apply to most same-sex relationships... or would that be "unconceivable"?
Few of us here are lawyers (I know I ain't), but aren't corporations considers "persons" for legal purposes?
"I, IBM, take thee, Red Hat, to be my lawful flunky..."
or even worse:
"I, Ben Affleck's Career, take thee, Jennifer Garner's Career, to be my legal co-producer, to have meetings with, schedule lunches, and arrange occasional carnal interludes with and between our respective Parties of the First Part...."
Gold has value because there is a standing market for it, in which people trade, and supply and demand have set a price. As the author of the paper states, gold is desirable because the supply is stable and the market is well-defined - it has industrial, cosmetic, and financial uses, all of which work to establish a price.
Platinum, he notes, is not seen as a "currency" because even though it's worth more per unit weight than gold (in the current market), the supply is unstable - new industrial uses for platinum are coming out all the time, and new sources and refining methods are constantly being worked. It's too unstable to be a worthwhile currency.
It's a "social Science" because they follow the scientific method as well as they can, but it's nigh unto impossible to actually perform meaningful experiments in economics above the micro- level (unless you're a world oil power, in which case you can experiment with demand-curves all you want). That's one of the weaknesses of the scientific method when applied to things above the biologcal scale - you can come up with the hypothesis, you can even come up with the experiment - but for it to be meaningful, you'd have to persuade a few thousand people to take part without modifying any other elements of their behavior. Tough for them.
This same guy was featured in one of the McGraw-Hill aviation mags (maybe Aviation Week?) preaching the same thing over 6 months ago - heck, I submitted it as a story then (Rejected, of course). He's watching his business base evaporate.
Green Hills can still make money if they debug their software, work on tools and add platforms - areas where F/OSS doesn't have the depth - esp. in niche processors.
You have one set of experts write the code under an F/OSS license, another set of experts examine the code, test cases, and test results.
Believe me, if you're talking about something like gunnery firmware, they're going to test it... the deepest fear in DoD these days is friendly fire.
The main cost driver in getting a PP-ASEL (I'm post-solo, so I know) is the cost of plane and instructor time. If this only requires 20 hours, versus 40 required for the Private Pilot, then that's a big stack of cash right there, esp. at $145/hr for dual time, $100/hr for solo in the C172-R I fly.
Wonder if my school is going to get any of these planes....
If the plane doesn't come from the manufacturer with specific equipment installed, the time and expense of the paperwork required to get approval to install new equipment, get the installation inspected and approved, and then get authorized to actually use said equipment in the air adds up quickly. This is non-trivial in the US (think of adding an IFR-capable GPS to an old piston twin to be used for charter service), but I understand it's nigh unto impossible in the EU right now.
Heh - powered lift is explicitly excluded from coverage under this rule.