Find some highly specialized programming skill and learn it. Employers are lot less picky if skill is not easily available on the market. For example now there is a lot of demand for Objective C coders for iPhone programming. You can get yourself a Mac, pick up some Objective C and move to iPhone then comfortable. You can start do some low-pay freelancing or pro bone coding as soon as you are able to code. That is of cause only example, I've no idea how long iPhone development boom to last. After getting some experience from the trenches you can get permanent employment more easy. Otherwise try to leverage your existing experience - accounting, statistics, marketing - whatever you are doing now. Write some simple app or utility which can help in your area.
Symbain C++ peculiarities is not a critical issue, they can be adapted to. But Symbian Signed is a real bummer, especially for small/indie developer. Pay for each attempt to sign binary + pay yearly for publisher ID. And self-signed application, not only limited in functionality, but will not be allowed to Nokia Ovi application store.
I don't think phones have the computing power to do dynamic features now
Actually they do. There are several working implementation of dense feature detectors on the smartphone, including SIFT, FAST and SURF, and some of them are capable of doing planar image registration(like Daniel Wagner "markerless" tracker. It's using image as fiduciary). The real difficulty is getting 3d structure form motion on the mobile, but that could be possible too soon. My opinion is tat 500+Mhz smartphone with floating point support should be capable of that, at least under good lighting conditions. The problem is to find correct algorithm with good enough optimization. Most of modern phones have DSP coprocessor, but API for it is closed by manufacturer. With open DSP API it would be practically sure thing.
I have tried Dwarf Fortress of cause, but in my opinion it doesn't go far enough. It generate world and allow to interact with previous player histories. What I mean is kind of Dwarf fortress with dozens of AI keeps and states running for hundreds seasons before player enter. Each state with reputation, agenda, religion and culture.
Really open game world should be procedurally generated IMHO, like roguelike games and their derivatives (diablo etc). The problem is that such world often look sterile and artificial, and need content created by designer to become more alive. I think the solution could be - after creating random seed world it should evolve for several hundreds generations. That way disbalances would die out, factions will have history of relationship, artifacts and places of power would have some logic in their placement. Kind of genetic algorithm for game content.
Experts want recognition of their efforts. It should not necessarily be money. It could be some kind of "elite" account, credits, reputation system etc. Of cause no expert would want his contribution be mutilated by opinionated teenager or some crank with agenda. The problem is how to identify experts. Some ID system like OpenID based on the university/corporate site/homepage or like could be useful.
I have a experience in cell phone marker tracking and IMO colored triangles are not the best solution. Colors are susceptible to lighting condition. This approach may have problem in poor lighting, where camera will have trouble to distinguish yellow from red and green from blue. I've tried colored markers and in my experience pure black and white work best. Also detection of rectangles is more reliable than detection of triangles/parallelograms under perspective transform. False positive for triangles also more likely.
Essentially the same idea (web camera is even in TFA, without microscope tough). Or may be only web camera sensor. After that make a count with some blob detector, or some more complex pattern recognition soft. Of cause PC+camera+microscope would cost 3-4 times more than just microscope, but you wouldn't need qualified medic to operate it, and it would be faster too.
Each of the Alexander the Great commanders was good or great military and political leader. But without Alexander they tear empire apart in several years. It usually take a strong and outstanding personality to successfully manage group of others outstanding personalities.
Perelman and Wiles both were working completely alone(or mostly alone for Wiles). Of cause they were using tools developed by others, but still they were lone researchers against insurmountable odds.
The OpenGL is harder mostly because it was developed for client-server architecture, with all the functions merely wrapping for the commands to server. That caused OpenGL states model to be quite complicated. As OpenGL evolved it got more and more states and extensions, some of which work little differently for different platforms/manufacturers. The plan was for OpenGL 3.0 to break legacy support, clean mess, get rid of states and introduce nice objects instead. However the task proved to be too difficult, and after three (or was it two?) years Khronos Group produced specification with very little improvements on the classic OpenGL model. Though the rumors saying that the reason of this was that the manufacturers (NVIDIA nad ATI) could not agree on standards.
You probably mistake dark energy for dark matter.
Dark energy is indeed most probably a cosmological constant and related to the energy of universe.
Dark matter is completely different thing - it's an invisible mass causing anomalous speed distribution of the galaxies in the clusters, stars in the galaxies and most spectacular - shape of the Bullet Cluster
Creativity and bipolar disorder was studied lately, and it seems there is some correlation. Cognitive deficits in bipolar disorder seems insignificant or absent. Generally mental disorder often associated with super-creative people, and not only artistic. Goedel and arguably Perelman come to mind.
It will work for all transforms, if of cause no three points are collinear and as I told surface is planar, which is the case. Math is trivial. Surface itself don't have to have distinctive features, enough if there are such features on its boundary. Straight edges itself could be tracked instead of points (projective duality). Lighting and color adjustment are not especially difficult if there is no CPU load limitation, there is a plenty of segmentation algorithms.
Augmented Reality isn't exactly a news, but talked about a lot lately. 3D tracking of planar surfaces from TFA is easiest kind of image registration, because planar surface 3d position could be restored from a single videoframe. All you have to do is identify 4+ permanent features on the surface, and you can consistently track it. In less calculating heavy cases, like rectangle tracking it could be done on the mobile phone. Feature tracking could be done on mobile too, but frame rate drop considerably.
From the gripping hand global warming begin turning huge expanses of northern taiga forests to wetlands. Aren't wetlands an ideal environment for amphibians ?
In that case lithography can help. Also similar methods like photoetching. And if you have money to burn, 3d scan and high-end 3d printing on durable substance.
Slashdot Mirror
And it may have some market. Range would increase and travel would make some exercise. Could be used by people who don't ride bicycle for some reason.
Find some highly specialized programming skill and learn it. Employers are lot less picky if skill is not easily available on the market. For example now there is a lot of demand for Objective C coders for iPhone programming. You can get yourself a Mac, pick up some Objective C and move to iPhone then comfortable. You can start do some low-pay freelancing or pro bone coding as soon as you are able to code. That is of cause only example, I've no idea how long iPhone development boom to last. After getting some experience from the trenches you can get permanent employment more easy. Otherwise try to leverage your existing experience - accounting, statistics, marketing - whatever you are doing now. Write some simple app or utility which can help in your area.
Symbain C++ peculiarities is not a critical issue, they can be adapted to. But Symbian Signed is a real bummer, especially for small/indie developer. Pay for each attempt to sign binary + pay yearly for publisher ID. And self-signed application, not only limited in functionality, but will not be allowed to Nokia Ovi application store.
Actually they do. There are several working implementation of dense feature detectors on the smartphone, including SIFT, FAST and SURF, and some of them are capable of doing planar image registration(like Daniel Wagner "markerless" tracker. It's using image as fiduciary). The real difficulty is getting 3d structure form motion on the mobile, but that could be possible too soon. My opinion is tat 500+Mhz smartphone with floating point support should be capable of that, at least under good lighting conditions. The problem is to find correct algorithm with good enough optimization. Most of modern phones have DSP coprocessor, but API for it is closed by manufacturer. With open DSP API it would be practically sure thing.
I have tried Dwarf Fortress of cause, but in my opinion it doesn't go far enough. It generate world and allow to interact with previous player histories. What I mean is kind of Dwarf fortress with dozens of AI keeps and states running for hundreds seasons before player enter. Each state with reputation, agenda, religion and culture.
Really open game world should be procedurally generated IMHO, like roguelike games and their derivatives (diablo etc). The problem is that such world often look sterile and artificial, and need content created by designer to become more alive. I think the solution could be - after creating random seed world it should evolve for several hundreds generations. That way disbalances would die out, factions will have history of relationship, artifacts and places of power would have some logic in their placement. Kind of genetic algorithm for game content.
Not all pdf are readable with foxit. I have foxit as default, but still have to use Acrobat from time to time.
Experts want recognition of their efforts. It should not necessarily be money. It could be some kind of "elite" account, credits, reputation system etc. Of cause no expert would want his contribution be mutilated by opinionated teenager or some crank with agenda. The problem is how to identify experts. Some ID system like OpenID based on the university/corporate site/homepage or like could be useful.
I have a experience in cell phone marker tracking and IMO colored triangles are not the best solution. Colors are susceptible to lighting condition. This approach may have problem in poor lighting, where camera will have trouble to distinguish yellow from red and green from blue. I've tried colored markers and in my experience pure black and white work best. Also detection of rectangles is more reliable than detection of triangles/parallelograms under perspective transform. False positive for triangles also more likely.
Essentially the same idea (web camera is even in TFA, without microscope tough). Or may be only web camera sensor. After that make a count with some blob detector, or some more complex pattern recognition soft. Of cause PC+camera+microscope would cost 3-4 times more than just microscope, but you wouldn't need qualified medic to operate it, and it would be faster too.
Each of the Alexander the Great commanders was good or great military and political leader. But without Alexander they tear empire apart in several years. It usually take a strong and outstanding personality to successfully manage group of others outstanding personalities.
Perelman and Wiles both were working completely alone(or mostly alone for Wiles). Of cause they were using tools developed by others, but still they were lone researchers against insurmountable odds.
I don't think so. For spammer 10% of success is a reliable CAPTCHA crack, but for OCR it's a failure.
The OpenGL is harder mostly because it was developed for client-server architecture, with all the functions merely wrapping for the commands to server. That caused OpenGL states model to be quite complicated. As OpenGL evolved it got more and more states and extensions, some of which work little differently for different platforms/manufacturers. The plan was for OpenGL 3.0 to break legacy support, clean mess, get rid of states and introduce nice objects instead. However the task proved to be too difficult, and after three (or was it two?) years Khronos Group produced specification with very little improvements on the classic OpenGL model. Though the rumors saying that the reason of this was that the manufacturers (NVIDIA nad ATI) could not agree on standards.
Really old Chinese kung-fu forms have songs associated with them. So they probably were mnemonic devices.
Evolutionary niche of Donald Duck ?
You probably mistake dark energy for dark matter. Dark energy is indeed most probably a cosmological constant and related to the energy of universe. Dark matter is completely different thing - it's an invisible mass causing anomalous speed distribution of the galaxies in the clusters, stars in the galaxies and most spectacular - shape of the Bullet Cluster
Any advices how to get more visibility for freeware (Nokia Symbian 9.x) ? I've tried Nokia Mosh, but it seems not doing much.
Creativity and bipolar disorder was studied lately, and it seems there is some correlation. Cognitive deficits in bipolar disorder seems insignificant or absent. Generally mental disorder often associated with super-creative people, and not only artistic. Goedel and arguably Perelman come to mind.
It will work for all transforms, if of cause no three points are collinear and as I told surface is planar, which is the case. Math is trivial. Surface itself don't have to have distinctive features, enough if there are such features on its boundary. Straight edges itself could be tracked instead of points (projective duality). Lighting and color adjustment are not especially difficult if there is no CPU load limitation, there is a plenty of segmentation algorithms.
Augmented Reality isn't exactly a news, but talked about a lot lately. 3D tracking of planar surfaces from TFA is easiest kind of image registration, because planar surface 3d position could be restored from a single videoframe. All you have to do is identify 4+ permanent features on the surface, and you can consistently track it. In less calculating heavy cases, like rectangle tracking it could be done on the mobile phone. Feature tracking could be done on mobile too, but frame rate drop considerably.
Wikipedia entry on the subject is more clear and concise. Also it's not exactly a news - wiki entry dates from four years ago.
From the gripping hand global warming begin turning huge expanses of northern taiga forests to wetlands. Aren't wetlands an ideal environment for amphibians ?
In that case lithography can help. Also similar methods like photoetching. And if you have money to burn, 3d scan and high-end 3d printing on durable substance.
Most of what we know of Shumer history is form their tax and business records.