For clients as graphically primitive as SecondLife, this is a relatively straightforward task of publishing a simple texture & mesh specification. But if you want to push things to support more complex graphics and more efficient avatar and object systems, you quickly step deep into implementation specific issues that generally kill efficiency across implementations.
I worked in games tools and engines for almost 20 years, with years of art path work and a focus on avatars and interoperability, and frankly the more efficient you design your system, the harder it is to describe it in simplistic generic terms. Add vested interests and committees, and you are likely to get a repeat of VRML - one company railroads the process to accept their spec, which hobbles progress forever.
Shameless plug: I've also been working for over a decade on massively multiplayer vr & games over p2p, something that will come online this year as proprietary, but move to open source once our small group leverages our first mover advantages. Our website doesn't show it yet, but the underlying tech is at least a generation past anything on the market to date - imagine a superset of Sims2 avatars and active objects with coding interfaces in Python and C++, in-engine collaborative editing of the world, open art import paths, integrated CreativeCommons rights, content rating, audio chat, all built on military grade crypto w/ Byzantine robustness. And we're always looking for more help, need more veteran programmers and human animators. http://www.vscape.com/
As long as the FCC props up access "right of way" monopolies, the free market cannot function. Between DSL distance constraints, spectrum auctions to the highest bidder, everybody overselling bandwidth, [nearly] everybody traffic shaping, unlimited service provider consolidation, and [nearly] every access provider requiring strict "you will be a consumer only" contracts, where is the free market? Net neutrality is just a bastion against unconstrained traffic shaping. The government has already sold off most of our other rights...
Some of us in the game industry have been using variations on these methods for years, I'm building a game now where most textures are procedural. Use of wavelets for this is nothing new either, I started playing with it myself ~ a decade ago, its always been the best approach for certain classes of image, especially for generating MIP levels. I hope the "new" idea here isn't trying to patent it...
One clarification w/r/t loading times, slow speeds, ect (simplified): With normal textures, you load your texture from disk, generally in one of the dds formats, then upload the desired subset of pregenerated MIP levels to texture memory. With procedural textures, you generate the texture (and the needed set of MIP levels), and upload them to texture memory. There is no need to regenerate and reload them unless someone decides to flush texture memory. The only time difference is disk load vs. generation, and you (generally) do that once.
From the article: Goldman and his team took human fetal midbrain tissues, in which dopamine cells are made, and extracted glial cells, whose normal role is to support and maintain the growth of neurons. They then cultured stem cells in this glia-rich environment.
I'm sure they have an professional ethecist on board who told them all is well, but I'd say this goes a wee bit beyond the use of stem cells harvested from blastocysts. Where exactly did they obtain "human fetal midbrain tissues"?
I cringe in disgust at how far this slippery slope is progressing...
There's plenty of private groups doing neat things...
<plug>
I run a small group, some of us game industry veterans, creating a MMO system over p2p. Cross platform front end w/ CgFX shaders over OpenGL or DirectX (= portable to next gen consoles), back end is hybrid of p2p with very few database and certificate authorities, all secured by very strong crypto. User content, fully scriptable (Lua now, Java soon, C++ if core group signs it), art paths for Max and Milkshape. Network includes IM, social network, bt based content and file exchange, voice chat, and micropayments with gateways back into traditional credit card processors. Plan to release as free for non-commercial use, most source will be opened (later.) And all development funded by the developers, who are tired of knowing how to do something really cool, and watching it destroyed by our previous employers (Accolade, EA, OnLive, and others...) We're getting ready to go public with demo, could use a few more experienced game coders, really need artists, but figure demo will attract all we need.:) email kerry at vscape.com if interested...
I find it amusing how the article attempts to claim the GPL isn't viral, using the following arguments: * Microsoft may have first called it that (therefore it must be false.) * The GPL doesn't infect other software on the same computer. (DUH?)
If I use any GPL code in my application, even one line, I have to release my application under the GPL license. Throw all the pointless qualifications you want around it, like only if I release the application to the public, and I can still release it under other licenses, and all the related conditions that likely apply to 1% of real world cases.
As a developer, I believe that developers have a right to make a living from their code. I also know that so few people donate for free software that most developers who rely on donations have to do something else to pay the bills. At the same time, I like to release 95% of my work as open source, so other developers can use it, some of which may also be trying to make a living as a developer. The Apache, BSD, even LGPL licenses handle that just fine.
Trying to spin the GPL as non-viral is foolish - the free software community should at least be honest about their agenda. I'm not saying its a bad agenda, it just happens to be incompatible with mine.
I found this fascinating - Einstein is an iconic figure, so criticism is not taken well, but I found these to be a fascinating read. No idea how good the underlying sources are, but if there is any merit to them, he may not deserve a good deal of the credit he is given. Reminds me of Tesla vs. Marconi or Tesla vs. Edison.
Why does questioning evolution result in answers like "everyone knows evolution is a fact, that's a stupid question."
There is a problem with science, and I think it begins with what science has become - "believe what we tell you is true", instead of "believe what you can prove."
When I look at the molecular biological paradoxes inherent in the evolution of the bacterial cilia into a flagellum, I think evolutionary biology involves more faith than belief in a god, even if that god is a "flying spaghetti monster".
The scientific method means creating hypothesis, running experiments to test your hypothesis, and being willing to discard your premise if you ran into even one experiment that invalidated your hypothesis.
Darwin himself acknowledged that his theory of evolution fails if it were ever possible to point to a single system in nature that could not be created through some linear sequence of mutations. Molecular biology now demonstrates MANY systems containing irreducible complexity. A good scientist should now be willing to question the validity of evolution.
Anyone who thinks that evolution is a "proven fact" needs to check the definition of "proven" and "fact".
There are no stupid questions, just stupid people...
I have to agree with the other posters here that this isn't about religion.
I see two problems:
The first is education - the crap that is called "science education" in the schools in this country is raising idiots. They are taught to regurgitate "facts", and the definition of "fact" has changed from "what is provable" to "what we tell you". Critical thinking is discouraged, experimentation has no lab budget, and standards are dropping wildly. I read once (can't find source) that several decades ago most middle school girls could tell you what an aileron was. Today I'd be surprised if more than a few percent of high school graduates have a clue.
The other problem is money and the absolute focus most entities (commercial and educational) now have on short-term profitability. Real science means taking risks, thinking about the long term, spending time on basic science so you can reap the rewards of understanding new principles, then discovering how they may be applied. Today any idea that looks unlikely to be signifigantly profitable within 18 months has almost no chance for funding. This is a good part of the reason why basic progress is stalling in most areas of science that do not have immediate commercial applications.
Fixing either of these requires fundamental changes in the mindset. Neither are likely to happen anytime soon, mostly for the same two reasons...
Companies today, especially the billion+ dollar ones (I work for one too), are only interested in making next quarters numbers for the stockholders. Mentoring implies they are going to continue to hire local engineers that you could mentor. What I see happening is companies are only hiring interns locally for grunt work while laying off the rest, while all but the most senior hires are in India and China, with the trend moving to China (why pay $20k for an engineer when you can pay $5k?) The job responsibilities of those remaining senior jobs can best be summarized as "make the crap we get back from India and China work." Your honest days work consists of "integrating" the outsourced work - read "debug and rewrite". Your mentoring will mostly consist of being available at 7am and 7pm 5+ nights a week for your daily phone conferences with the outsourced teams, mostly reexplaining the specifications you sent, and pointing out the hundreds of ways their last deliverable failed to met even the most basic entrance criteria. In the meantime, you and the remaining interns will talk about how they should just hire a few more locals and let you finish the project yourself.
This is the new economy, and how high-tech works for the foreseeable future - everyone I know that still has a job is being reamed by their employer, including the ones on the "best companies to work for" lists. If you want something honest, start your own company and do it right.
ASN.1 is a truly poor specification. Yes, it sort of works in several spaces, I use them as part of my job reading and writing various credentials. But the spec is ambiguous, allows creation of documents that break parsers, and cannot be left-right parsed cleanly.
I suggest IFF. I've been working on an open spec for 3d content based on XML as a hobby project since VRML dropped the ball, and at the same time working on a parallel IFF chunk hierarchy based spec.
This lets me have text XML when I need human readable, and lets me have quickly parsed binary data when I need that - best of both worlds, and trivial to translate between. From experimenting with all the ways to do this, I've found XML/IFF chunking to be a clean map...
I'm still getting calls from companies in the Bay Area. The problem is that none of them pay enough to support a family outside of a slum. $80-100k is decent, but not when a "fixer-upper" costs $500k and needs $100k in repairs to keep it from falling down when you slam the door.
Actually, "well regulated" at the time the second ammendment was written would translate today into "well maintained and accurate". Timepieces and tools of the same era were also marketed as "well regulated", meaning they kept time or performed their job well.
The concept that this has anything to do with organized militias was introduced by groups trying to distort the meaning so they could 'regulate' in the modern sense, i.e. pass gun control laws.
By a strict reading of the 2nd ammendment, ALL modern US laws "regulating" firearms (and most other weapons) are unconstitutional.
They mean the lock is resistant to using an x-ray machine to peek at the pin / disc positions inside, usually by using something like lead around the cylinder / wheels.
If you want a really pick resistant mechanical lock, I suggest Medeco.
That's the public myth. The reality is different. If you have a juvenile criminal record, it may be sealed from the public, but it is not purged from the court system. Unless you fight to have it purged, it can be used against you as an adult. I have a friend who learned this the hard way.
Recovery is a big problem for high power rocketry - how many hours downrange do you want to drive to get your rocket back?
Normal approach is to eject a small drogue parachute near the apex of the flight, which is intended too let you lose lots of altitude without going to fast. It also keeps you from drifting too far from the launch site. Once you're close to the ground (via redundant altimeters) you eject the main chute to set it down 'relatively' gently.
"They argue that many water shortages could simply be solved by better conservation of existing supplies."
As long as population continues to increase, conservation and other increases in efficiency are only short-term solutions. Sooner or later you MUST increase the supply, or you run out.
Look at California's electricity problems for a good example of where this leaves you.
When I quit Cisco, I was the only real security programmer left in my business unit - all the other positions had been "outsourced" to Bangalore. That team didn't write "bad" code, it just wasn't robust. And they didn't get it. And management didn't care. And marketing just wants it to ship with the feature checklist complete.
I said it below, I'll say it again here. Companies have to CARE enough about security to have experienced crypto people do this sort of work. To design it, to implement it, and to test it.
This is yet another example of why you need to hire security programmers with actual experience in the field, not just outsource it to a cheap Indian programming group with no real experience writing robust protocols.
I'm an ex Cisco security programmer, and thats exactally what was happening before I quit. I wish I could say more...
I maintain at home a library of code under an open source (BSD style) license. Once an employer has given me permission to use and contribute back into that library, THEN I use it at work, and I take it with me when I go. If they want to pay me to reinvent the wheel, its on their dime, no problem! If they want the advantages my existing code, then they have to let me keep improvements. Sort of an informal LGPL license, without the "lesser" wording that smacks of posturing.
To the point - This survey did not allow me to clarify that my code library brought between jobs was legal, and it bugs me that it may be used to support the position that more people ignore the law than may actually do so.
As someone that just left the game industry for the second time - this time 'cause I got tired of looking for a job that didn't require a minimum of 80 hour weeks working on either a n-th generation sequel or a game that tried to differentiate itself through pushing the gore / splatter level:
The problem is with the people FUNDING the game industry. The independent shops are being swallowed by companies that have made loads of cash getting away with pumping out sequels that have only minor engine improvements. This sucks, but worked for a while in a few profitable genres. Many companies that tried to push it died after too many generations (I used to work at Accolade, that's part of what killed them...)
Unfortunately, people that funds games look at this seeming no-risk model, and refuse to fund anything that doesn't look like the same. They all want you to license an existing engine, and make a game that can be described in a single sentence as {profitable game A } crossed with {profitable game B.}
If you don't follow this model, you don't get funds.
As a related point, there are WAY too many companies in the industry for the amount of shelf space available, and the big players BUY shelf space, so its nearly impossible to compete anyway without cutting a deal with an existing major distributor. Want to do that? Guess what, you have to change your game to follow the same model as everyone else.
In the mean time, the EA's and Sony's of the world are pushing their developers harder and harder - they've currently got a surplus of available headcounts to replace all the burnt out ones with...
The industry needs more "angel" funders. But in this economy...
The problem starts at management...
on
Exploiting Software
·
· Score: 5, Informative
I write security software for a living, occasionally. I'm back in it now, after leaving it for a while out of frustration, among other things.
The problem is that even for us security geeks, its nearly impossible to get management to buy into spending the time to make things secure. Nobody would believe the stories I could tell about having security features gutted 'cause marketing decided certificates were too complicated, or 'cause access control systems were too hard to use, so "just put the built in passwords back in", or "*** wants a copy of the private keys", or "*** told me to tell you to check the private keys into version control". Stupid, stupid...
Until companies start caring about making products that are ACTUALLY secure, instead of just hiring security geeks to act as figureheads, then not letting them do their jobs, then system will continue to get hacked.
Slashdot Mirror
For clients as graphically primitive as SecondLife, this is a relatively straightforward task of publishing a simple texture & mesh specification. But if you want to push things to support more complex graphics and more efficient avatar and object systems, you quickly step deep into implementation specific issues that generally kill efficiency across implementations.
I worked in games tools and engines for almost 20 years, with years of art path work and a focus on avatars and interoperability, and frankly the more efficient you design your system, the harder it is to describe it in simplistic generic terms. Add vested interests and committees, and you are likely to get a repeat of VRML - one company railroads the process to accept their spec, which hobbles progress forever.
Shameless plug: I've also been working for over a decade on massively multiplayer vr & games over p2p, something that will come online this year as proprietary, but move to open source once our small group leverages our first mover advantages. Our website doesn't show it yet, but the underlying tech is at least a generation past anything on the market to date - imagine a superset of Sims2 avatars and active objects with coding interfaces in Python and C++, in-engine collaborative editing of the world, open art import paths, integrated CreativeCommons rights, content rating, audio chat, all built on military grade crypto w/ Byzantine robustness. And we're always looking for more help, need more veteran programmers and human animators. http://www.vscape.com/
As long as the FCC props up access "right of way" monopolies, the free market cannot function. Between DSL distance constraints, spectrum auctions to the highest bidder, everybody overselling bandwidth, [nearly] everybody traffic shaping, unlimited service provider consolidation, and [nearly] every access provider requiring strict "you will be a consumer only" contracts, where is the free market? Net neutrality is just a bastion against unconstrained traffic shaping. The government has already sold off most of our other rights...
Some of us in the game industry have been using variations on these methods for years, I'm building a game now where most textures are procedural. Use of wavelets for this is nothing new either, I started playing with it myself ~ a decade ago, its always been the best approach for certain classes of image, especially for generating MIP levels. I hope the "new" idea here isn't trying to patent it...
One clarification w/r/t loading times, slow speeds, ect (simplified): With normal textures, you load your texture from disk, generally in one of the dds formats, then upload the desired subset of pregenerated MIP levels to texture memory. With procedural textures, you generate the texture (and the needed set of MIP levels), and upload them to texture memory. There is no need to regenerate and reload them unless someone decides to flush texture memory. The only time difference is disk load vs. generation, and you (generally) do that once.
From the article: Goldman and his team took human fetal midbrain tissues, in which dopamine cells are made, and extracted glial cells, whose normal role is to support and maintain the growth of neurons. They then cultured stem cells in this glia-rich environment.
I'm sure they have an professional ethecist on board who told them all is well, but I'd say this goes a wee bit beyond the use of stem cells harvested from blastocysts. Where exactly did they obtain "human fetal midbrain tissues"?
I cringe in disgust at how far this slippery slope is progressing...
There's plenty of private groups doing neat things...
:) email kerry at vscape.com if interested...
<plug>
I run a small group, some of us game industry veterans, creating a MMO system over p2p. Cross platform front end w/ CgFX shaders over OpenGL or DirectX (= portable to next gen consoles), back end is hybrid of p2p with very few database and certificate authorities, all secured by very strong crypto. User content, fully scriptable (Lua now, Java soon, C++ if core group signs it), art paths for Max and Milkshape. Network includes IM, social network, bt based content and file exchange, voice chat, and micropayments with gateways back into traditional credit card processors. Plan to release as free for non-commercial use, most source will be opened (later.) And all development funded by the developers, who are tired of knowing how to do something really cool, and watching it destroyed by our previous employers (Accolade, EA, OnLive, and others...) We're getting ready to go public with demo, could use a few more experienced game coders, really need artists, but figure demo will attract all we need.
</plug>
I find it amusing how the article attempts to claim the GPL isn't viral, using the following arguments:
* Microsoft may have first called it that (therefore it must be false.)
* The GPL doesn't infect other software on the same computer. (DUH?)
If I use any GPL code in my application, even one line, I have to release my application under the GPL license. Throw all the pointless qualifications you want around it, like only if I release the application to the public, and I can still release it under other licenses, and all the related conditions that likely apply to 1% of real world cases.
As a developer, I believe that developers have a right to make a living from their code. I also know that so few people donate for free software that most developers who rely on donations have to do something else to pay the bills. At the same time, I like to release 95% of my work as open source, so other developers can use it, some of which may also be trying to make a living as a developer. The Apache, BSD, even LGPL licenses handle that just fine.
Trying to spin the GPL as non-viral is foolish - the free software community should at least be honest about their agenda. I'm not saying its a bad agenda, it just happens to be incompatible with mine.
Some of us are... :)
http://www.pbs.org/opb/einsteinswife/
r/e Mileva Maric
I found this fascinating - Einstein is an iconic figure, so criticism is not taken well, but I found these to be a fascinating read. No idea how good the underlying sources are, but if there is any merit to them, he may not deserve a good deal of the credit he is given. Reminds me of Tesla vs. Marconi or Tesla vs. Edison.
Yes:
Why is the theory of evolution taught as a fact?
Why does questioning evolution result in answers like "everyone knows evolution is a fact, that's a stupid question."
There is a problem with science, and I think it begins with what science has become - "believe what we tell you is true", instead of "believe what you can prove."
When I look at the molecular biological paradoxes inherent in the evolution of the bacterial cilia into a flagellum, I think evolutionary biology involves more faith than belief in a god, even if that god is a "flying spaghetti monster".
The scientific method means creating hypothesis, running experiments to test your hypothesis, and being willing to discard your premise if you ran into even one experiment that invalidated your hypothesis.
Darwin himself acknowledged that his theory of evolution fails if it were ever possible to point to a single system in nature that could not be created through some linear sequence of mutations. Molecular biology now demonstrates MANY systems containing irreducible complexity. A good scientist should now be willing to question the validity of evolution.
Anyone who thinks that evolution is a "proven fact" needs to check the definition of "proven" and "fact".
There are no stupid questions, just stupid people...
I have to agree with the other posters here that this isn't about religion.
I see two problems:
The first is education - the crap that is called "science education" in the schools in this country is raising idiots. They are taught to regurgitate "facts", and the definition of "fact" has changed from "what is provable" to "what we tell you". Critical thinking is discouraged, experimentation has no lab budget, and standards are dropping wildly. I read once (can't find source) that several decades ago most middle school girls could tell you what an aileron was. Today I'd be surprised if more than a few percent of high school graduates have a clue.
The other problem is money and the absolute focus most entities (commercial and educational) now have on short-term profitability. Real science means taking risks, thinking about the long term, spending time on basic science so you can reap the rewards of understanding new principles, then discovering how they may be applied. Today any idea that looks unlikely to be signifigantly profitable within 18 months has almost no chance for funding. This is a good part of the reason why basic progress is stalling in most areas of science that do not have immediate commercial applications.
Fixing either of these requires fundamental changes in the mindset. Neither are likely to happen anytime soon, mostly for the same two reasons...
Companies today, especially the billion+ dollar ones (I work for one too), are only interested in making next quarters numbers for the stockholders. Mentoring implies they are going to continue to hire local engineers that you could mentor. What I see happening is companies are only hiring interns locally for grunt work while laying off the rest, while all but the most senior hires are in India and China, with the trend moving to China (why pay $20k for an engineer when you can pay $5k?) The job responsibilities of those remaining senior jobs can best be summarized as "make the crap we get back from India and China work." Your honest days work consists of "integrating" the outsourced work - read "debug and rewrite". Your mentoring will mostly consist of being available at 7am and 7pm 5+ nights a week for your daily phone conferences with the outsourced teams, mostly reexplaining the specifications you sent, and pointing out the hundreds of ways their last deliverable failed to met even the most basic entrance criteria. In the meantime, you and the remaining interns will talk about how they should just hire a few more locals and let you finish the project yourself.
This is the new economy, and how high-tech works for the foreseeable future - everyone I know that still has a job is being reamed by their employer, including the ones on the "best companies to work for" lists. If you want something honest, start your own company and do it right.
I suggest IFF. I've been working on an open spec for 3d content based on XML as a hobby project since VRML dropped the ball, and at the same time working on a parallel IFF chunk hierarchy based spec.
This lets me have text XML when I need human readable, and lets me have quickly parsed binary data when I need that - best of both worlds, and trivial to translate between. From experimenting with all the ways to do this, I've found XML/IFF chunking to be a clean map...
An older version of the spec is at http://www.vscape.com/vml/index.html, if anyone wants an example of how this could work.
I'm still getting calls from companies in the Bay Area. The problem is that none of them pay enough to support a family outside of a slum. $80-100k is decent, but not when a "fixer-upper" costs $500k and needs $100k in repairs to keep it from falling down when you slam the door.
Actually, "well regulated" at the time the second ammendment was written would translate today into "well maintained and accurate". Timepieces and tools of the same era were also marketed as "well regulated", meaning they kept time or performed their job well.
The concept that this has anything to do with organized militias was introduced by groups trying to distort the meaning so they could 'regulate' in the modern sense, i.e. pass gun control laws.
By a strict reading of the 2nd ammendment, ALL modern US laws "regulating" firearms (and most other weapons) are unconstitutional.
They mean the lock is resistant to using an x-ray machine to peek at the pin / disc positions inside, usually by using something like lead around the cylinder / wheels.
If you want a really pick resistant mechanical lock, I suggest Medeco.
That's the public myth. The reality is different. If you have a juvenile criminal record, it may be sealed from the public, but it is not purged from the court system. Unless you fight to have it purged, it can be used against you as an adult. I have a friend who learned this the hard way.
Recovery is a big problem for high power rocketry - how many hours downrange do you want to drive to get your rocket back?
Normal approach is to eject a small drogue parachute near the apex of the flight, which is intended too let you lose lots of altitude without going to fast. It also keeps you from drifting too far from the launch site. Once you're close to the ground (via redundant altimeters) you eject the main chute to set it down 'relatively' gently.
Just a quick comment on the toxicity vs. medicinal value of Lithium:
Common "therapeutic" dosage of Lithium is 14-28 mg/kg/day.
LD50 (kills 50%) is officially around 710 mg/kg/day.
However, permanent neurological damage has been documented at levels as low as 170 mg/kg/day.
Not exactly a safe substance. A friend once took a weeks supply of his Lithium in a suicide attempt. Left him severely retarded.
As long as population continues to increase, conservation and other increases in efficiency are only short-term solutions. Sooner or later you MUST increase the supply, or you run out.
Look at California's electricity problems for a good example of where this leaves you.
I was there. It is due to outsourcing. Period.
When I quit Cisco, I was the only real security programmer left in my business unit - all the other positions had been "outsourced" to Bangalore. That team didn't write "bad" code, it just wasn't robust. And they didn't get it. And management didn't care. And marketing just wants it to ship with the feature checklist complete.
I said it below, I'll say it again here. Companies have to CARE enough about security to have experienced crypto people do this sort of work. To design it, to implement it, and to test it.
But now its all about keeping things cheap.
This is yet another example of why you need to hire security programmers with actual experience in the field, not just outsource it to a cheap Indian programming group with no real experience writing robust protocols.
I'm an ex Cisco security programmer, and thats exactally what was happening before I quit. I wish I could say more...
I maintain at home a library of code under an open source (BSD style) license. Once an employer has given me permission to use and contribute back into that library, THEN I use it at work, and I take it with me when I go. If they want to pay me to reinvent the wheel, its on their dime, no problem! If they want the advantages my existing code, then they have to let me keep improvements. Sort of an informal LGPL license, without the "lesser" wording that smacks of posturing.
To the point - This survey did not allow me to clarify that my code library brought between jobs was legal, and it bugs me that it may be used to support the position that more people ignore the law than may actually do so.
As someone that just left the game industry for the second time - this time 'cause I got tired of looking for a job that didn't require a minimum of 80 hour weeks working on either a n-th generation sequel or a game that tried to differentiate itself through pushing the gore / splatter level:
The problem is with the people FUNDING the game industry. The independent shops are being swallowed by companies that have made loads of cash getting away with pumping out sequels that have only minor engine improvements. This sucks, but worked for a while in a few profitable genres. Many companies that tried to push it died after too many generations (I used to work at Accolade, that's part of what killed them...)
Unfortunately, people that funds games look at this seeming no-risk model, and refuse to fund anything that doesn't look like the same. They all want you to license an existing engine, and make a game that can be described in a single sentence as {profitable game A } crossed with {profitable game B.}
If you don't follow this model, you don't get funds.
As a related point, there are WAY too many companies in the industry for the amount of shelf space available, and the big players BUY shelf space, so its nearly impossible to compete anyway without cutting a deal with an existing major distributor. Want to do that? Guess what, you have to change your game to follow the same model as everyone else.
In the mean time, the EA's and Sony's of the world are pushing their developers harder and harder - they've currently got a surplus of available headcounts to replace all the burnt out ones with...
The industry needs more "angel" funders. But in this economy...
I write security software for a living, occasionally. I'm back in it now, after leaving it for a while out of frustration, among other things.
The problem is that even for us security geeks, its nearly impossible to get management to buy into spending the time to make things secure. Nobody would believe the stories I could tell about having security features gutted 'cause marketing decided certificates were too complicated, or 'cause access control systems were too hard to use, so "just put the built in passwords back in", or "*** wants a copy of the private keys", or "*** told me to tell you to check the private keys into version control". Stupid, stupid...
Until companies start caring about making products that are ACTUALLY secure, instead of just hiring security geeks to act as figureheads, then not letting them do their jobs, then system will continue to get hacked.