I'm probably stepping in shit here (and I happen to HATE Verizon), but I need to understand your argument better. >>>they POCKETED the BILLIONS of dollars they got from the government to make sure that fiber rollout occured [sic] Are you talking about the money they earn as a result of their protected monopoly position? Verizon and other RBOCs are under no statutory obligation to run fiber to everyone's home. Are you saying that they have received taxpayer funds in order to do this?
What I'm saying is that we can't rely on the RBOCs to provide any kind of fast, cheap DSL service because it's not in their interest to do so. And that means someone else needs to step up and do it. The FCC's action in effect means (for reasons having nothing to do with technology) that DSL will not be the technology that meets this need.
I'm not sure about the underlying reasoning here- FCC has changed a lot since Mike Powell left. They may have decided that preventing anticompetitive behavior is too much intereference in free markets, although that hasn't often stopped them before.
But it is obvious that this move will constrain DSL's quality and price advantages. I have to suffer with Verizon and I'm in PAIN everytime I use it and everytime I see my bill. Things will only get worse. What the hell is everyone waiting for? Oh right, I forgot, you're waiting for SOMEONE ELSE to take the risk.
Every other country in the world has far better broadband infrastructure than the US, and it's because of the behavior of our monopoly providers.
I hope this will finally provide an incentive for metropolitan wifi providers and cable vendors to decide upon and implement a reasonable business model. Guys, it isn't rocket science, and it's not going to take anything like the irrational capital-deployment decisions that resulted in the long-haul-fiber overbuild.
One more time: it will take some significant engineering. The old escape towers were just rocket-powered ejection seats with explosive bolts. That technology is not going to get you an intact and fully-functioning crew vehicle that can survive an unpowered fall from 200,000 feet back to an arbitrary spot in the Atlantic Ocean.
IIRC, the Mercury/Gemini/Apollo escape towers were engineered primarily to deal with problems on liftoff. It's not too many minutes into the flight that you're 1000 miles downrange and way out of the breathable atmosphere. What then? Serious improvements are clearly needed, even if the basic design is workable. Also, the early towers were very very dangerous- the few times they went off in tests and simulations, they did an awful lot of damage. They were really a last-ditch hope.
I'll buy the analysis which holds the solid-fuel rocket is far more reliable, and of course in a space program where manned launches are "routine," the payoff is huge.
But now you have to engineer a crew-escape mechanism in case of a serious problem during the boost phase. Can we improve on the Mercury-era "escape tower," basically a rocket-powered ejection seat? We need to enable a safe, reliable crew recovery at any point from liftoff till the solid booster burns out, which occurs at a significant altitude and distance downrange.
Notice that the story is about a guy with a partisan interest in the outcome- he's a Thiokol engineer and they make the solid-fuel boosters for the current shuttle. These boosters are the heart of the proposal, and my only question is, do we want solid fuel rockets as the primary lifter for human crews? Don't they present special challenges and risks because they can't be shut down in case of problems? Just asking, IANA astronautical engineer.
Apart from that, this seems like a good blending of proven tech from the shuttle project with a more clear-eyed view of overall project goals, with favorable economics.
The original poster asks, "can't we do better"? As long as we are relying on combustion of chemical fuels, I don't think there's any need to do "better." A quantum leap forward in lift technology will have to await a new type of propulsion, which is at least decades away. In the meantime, let's get busy!
I'm a holder of several software patents, both issued and pending. I vehemently disagree with the whole concept of software patents, but in today's broken environment, you have to apply for these patents as a defensive measure. Otherwise you may find that someone else can prevent you from commercializing your own ideas.
And having been through the process, I can tell you that most examiners at the USPTO are simple, unimpressive bureaucrats with nowhere near the capability and insight needed to evaluate software innovations. They are not operating in bad faith, they just aren't equipped to do the job that the big shots in our industry are asking of them. (Further proof of this point is that they usually retire as soon as they are eligibile, and ever-increasing numbers of them leave the service early. It's a crummy job.)
The path of least resistance for the PTO is to just grant about 60% of the applications put before them and hope that the courts will sort out the conflicts. That's a great solution for the small technology shops and individuals that produce the real innovations, isn't it? Especially since you're now relying on an equally ignorant judge and jury, who will be favorably impressed by those nice, polished, well-dressed attorneys from IBM/Microsoft/Oracle/Toshiba/whoever.
LEGISLATION is needed to solve this problem. The community needs to draft a revision of the patent laws and lobby its passage through Congress. Anybody interested in taking up the challenge?
This was in summer of 1983. I'd been writing Apple and PC programs for years in 6502 and 8088 assembly. I couldn't believe the $10,000 price tag, which was so outlandish you'd never even think of recommending it to clients (a 20meg Corvus hard disk was $3,500 or so at that time, iirc). But the bitmapped GUI was unlike anything you'd ever imagined. It wasn't even obvious how to start working with the thing. No one else was talking about anything similar at that time, at least not for practical, widespread usage. I think you're wrong about the hundred companies scrambling to implement it. It was one of those ideas whose brilliance took quite a while to become apparent to most people. About a year later I started hearing rumors out of Redmond about something totally new called "event-driven programming," and of course we know that MSFT dates the original creation of Windows to 1983, with 1.0 shipping in 1985.
It's incredible how much ignorance has been spewed on this thread. What makes the American system different from all others is our emphasis on freedom. Economic systems derive power directly from the vitality and social cohesion of the societies which generate them. Free societies with high social cohesion generate the most economic value and power because all the incentives align. American society is losing its social cohesion, mostly because so many people believe all the idiocy being spewed on this thread and everywhere else. China's government is very successfully motivating huge sectors of its subject population with nationalism and the dream of recovering past greatness. This strategy is profoundly dehumanizing, but it generates tremendous national purpose and cohesion. The last government that tried it was Nazi Germany, and it worked so well there that the rest of the world had to stop them by force.
We like to think that free societies are happy and successful because they are free and open, and in fact the example of their success will encourage others. China is now trying to prove the opposite. They are trying to control their own people, and motivate them through a shared sense of national purpose and recovery of past greatness. The last government that tried this was the Nazis. And it took millions of lives to suppress that threat. The government of China is replaying the experiment. But they have time, numbers, capital, and unlimited reserves of patience on their side. We are now engaged in the last great test of freedom, people. Wake up, we live in interesting times.
A lot of what is being said about this pair of niche OS's is reminiscent of what was said about Linux back when it was new. A lot of new things get tried, but very very manage to put together that magic combination of meeting a real need with sustained, competent execution. That means that, as silly as some of you will think, the next big phenomenon comparable to Mosaic or Linux is just getting off the ground now. The few people becoming aware of it now are saying things like "hell, no driver support or PPP? COME ON!" I'm not saying it's either Syllable or SkyOS but it will certainly be something. And here's the big need that everyone has: A COMPETENT, FREE DESKTOP O/S. Windows sucks, OS X is expensive, and Linux is failing on the desktop for a number of reasons, both good and bad. The next big thing will meet this need.
Nah, the late Eighteenth is still the Enlightenment. (Unless you're counting The Sorrows of Young Werther, which would be defensible.) Also, philosophical currents tend to reach music two or three decades after they affect literature and the visual arts. More directly to your point, think about Schopenhauer's statements on art. That gets you to into the early Nineteenth. I've never liked that term "Romantic" anyway- too overloaded. To me, Romanticism is primarily about that notion that the human spirit finds deep echoes in the natural world. If you accept that, then the "Romantic" Age doesn't really start until well into the Nineteenth.
Improvisation may be an interesting way of producing pieces of immediate performance art in small forms. And it was considered an essential skill in music pedagogy through the eighteenth century. But it's the wrong way to write larger forms like symphonies. Large software systems are more like symphonies than jazz-sets. They require a lot of long-range planning. The great symphonists of the nineteenth and early twentieth centuries had an incredible gift for long-range thinking, combined with an unbelieveable ability to keep all of a large production in their heads. Here's a possible clue to the required mental equipment: there are anecdotes about most of the the great composers regarding their prodigious, almost inhuman memory-power.
What would Mozart have thought?
on
Is Programming Art?
·
· Score: 4, Insightful
Mozart considered composition a craft. So did Bach, who regularly turned out a new cantata most weeks for his job at the Thomaskirche in Leipzig. The notion that artists have special access to some emotional content not available to ordinary craftsmen is a nineteenth-century idea. But everyone agrees that both Mozart and Bach had access to some pretty unusual stuff- we hear it and respond to it. The content of programming is perhaps too instrumental (i.e., interesting for its usefulness more than its inherent qualities) to rise to the level of art. But this may be changing with the state-of-the-art games. In a hundred years, people may look back at today's game developers as the inventors of a new art form!
Good point, although 9/11 (more than a year after the bust) was the trigger that destroyed airline equity. Thanks to yield management and the capital-glut, the 1990s was an extremely good decade for the airlines. Remember paying $2500 several times a month for the New York-San Francisco milk run? I sure do;-)
...and it ain't coming back. What we are seeing is a regression to the mean. In the Internet days, major enterprises with traditional business models watched in horror as Amazon appeared to make the whole world obsolete. Priceline, nothing more than a travel agent, at one time had a higher market cap than all the major US airlines combined. This was a historic misallocation of capital, comparable to the overinvestment in radio in the 1920s. Every CEO in America asked himself whether a pimply geek like Jeff Bezos could blow up out of nowhere and destroy his whole business just like that. We know how stupid that was in hindsight, but fear is an incredibly powerful motivator. The result was a tremendous amount of IT overspending. And for those of us who were right in the middle of it, it was rational at the time!
China is finally getting around to doing in software what they have always done in industries deemed "strategic" - they force world-leading companies to share technology with Chinese partners and eventually compete against the foreigners. It works because Western businesses feel like they need exposure to China's purported "billion-person market" at any cost. Look at Boeing. Look at Cisco. Now it's true that the economic impact in China is negligible because they don't buy software from the West at all - they steal it. And TFA quotes the USTA pathetically whining about how China is closing themselves out of the world's best software. WAKE UP, you fucking morons in the US Dept of Commerce: China wants the "best software in the world" to be made in China, not Palo Alto, not Redmond, not Bangalore, not anywhere else. This is the first shot in a trade war aimed right at one of the few industries left where the US can honestly be said to have a technological edge.
The Register's take on this (quoted above) indicates that MSFT slashed R&D spending by half, from about $2.5B in last year's Q3 to about $1.5B this year. This can't possibly be right unless their R&D expenditures are not uniform across the year. The $1.4B number is a bit on the low side, you expect a license-based software company to spend between 15% and 18% of revenue on R&D. But $2.5B on $10B in revenue is way the hell out there, just can't be right. Anyone know what the Register's source was?
And BTW, it's striking that Microsoft (or anyone) can spend $5B a year on some of the best programmers in the world, and not get anything new and different to show for it. That's why they have so much cash in the bank: they have nothing to spend it on. That's also why Microsoft's owners (that would be us, the investing public) have been clamoring for the cash to be returned to them in the form of special dividends.
One poster said correctly that, out of 10 deals, a VC looks for 1-2 home runs, maybe 3-4 breakevens and the rest are total losses. What smart VCs do is take a perspective on the market as a whole and bet on what are the coming hot segments. Then they carefully place a few bets in those spaces. To be chosen, a company has to have a top management team, be focused (or re-focusable) on the laser-narrow segment of the market that corresponds to the VC's view, have a "correct" business model, and can check off a whole list of other variables. Everything about the company has to right, down to what color ties the CEO wears (if any). Basically, a VC manages his risk by only choosing companies that meet a whole range of very narrow constraints, with the only degree of freedom being the specific market segment, and that is chosen by the VC. This year, the VCs' tea leaves are showing open source as the hot space. One very interesting comment in TFA was the initial reaction to Fleury's attempts to get funded four years ago: "you must be nuts." Since he didn't fit the VCs' pre-established business-model checkbox at that time, he couldn't get funded. The VC view of the world has changed, and now the "open source" aspect is the hot one. Another thing good VCs always do is fund to milestones. If you don't hit substantially all of your targets, they will ruthlessly shoot you in the head and not fund your next round. This will either win big for the early VCs or it will fail. We'll know in about a year (that's a typical length for a funding round).
I looked through a 50,000 line C++ program I wrote last year. There are 83 goto statements- I was damn surprised there were so many! An average of 8 goto's occur in each of ten different functions. Each of the functions is a large state-machine processor (they are actually PDU handlers for network protocols), so in each case the basic function structure is about the same. The goto's are there to break out of the state machine after detecting an error. Each error is completely handled where it occurs but of course invalidates the transaction, which cannot continue. So the goto is actually more natural in this specific case than an exception handler.
If you're going to (judiciously) use goto where appropriate in C++ programs, note one important thing: a goto may not span automatic variable declarations. This should make sense, since the goto target wouldn't how know to reset the stack-frame pointer. This restriction doesn't apply in C, which only allows automatic declarations at the top a function.
There's a presumption built into this whole discussion that the human race is summum valorem, and must be preserved at all costs. I counterpropose that human beings are more valuable than the human race. It just might be the height of arrogance to suppose that we should do everything in our considerable power to colonize space simply for the purpose of perpetuating our own spawn, rather than solving the many problems we face here. The proverbial Martian would be well-justified in wanting to prevent a race intent on committing nuclear suicide from spreading over the Universe! Of course, if we really are alone in the universe, and in fact nothing but a coincidence (in itself an anti-scientific view), then my argument is false.
They counterfeit $100 bills. Fake American hundreds were "legal" tender on Russian black markets for a long while. Fake US currency printed in Lebanon looks better than the real thing, and often appears before the real stuff does too.
Hibernate's fans say it's the "right" way to handle extremely large, complex data models, especially if they already exist. What you have to do is write yourself a new Java class (according a to big complex set of rules), one for each entity in the model. And there are a lot of different ways to handle various association-types and cascading-persistence requirements. Once you've done all this, you have Hibernate's methods for "easily" loading and saving the objects. So in a way you have to re-express an already-existing data model in code. Anybody with experience in the tool care to argue whether it's worth the effort?
>>>You asked a person apiece at several F500 firms, and think you got an accurate picture of open-source's deployment? No, I don't think that. Since these people were CIOs or their direct reports, I think I got a picture of how F/OSS deployments are viewed at the top of the food chain. These are the people who are going to shut down all the F/OSS-based skunk-works projects as soon as they start getting important, if they don't like the answers about where support is coming from.
Slashdot Mirror
I'm probably stepping in shit here (and I happen to HATE Verizon), but I need to understand your argument better.
>>>they POCKETED the BILLIONS of dollars they got from the government to make sure that fiber rollout occured [sic]
Are you talking about the money they earn as a result of their protected monopoly position? Verizon and other RBOCs are under no statutory obligation to run fiber to everyone's home. Are you saying that they have received taxpayer funds in order to do this?
What I'm saying is that we can't rely on the RBOCs to provide any kind of fast, cheap DSL service because it's not in their interest to do so. And that means someone else needs to step up and do it. The FCC's action in effect means (for reasons having nothing to do with technology) that DSL will not be the technology that meets this need.
I'm not sure about the underlying reasoning here- FCC has changed a lot since Mike Powell left. They may have decided that preventing anticompetitive behavior is too much intereference in free markets, although that hasn't often stopped them before.
But it is obvious that this move will constrain DSL's quality and price advantages. I have to suffer with Verizon and I'm in PAIN everytime I use it and everytime I see my bill. Things will only get worse. What the hell is everyone waiting for? Oh right, I forgot, you're waiting for SOMEONE ELSE to take the risk.
Every other country in the world has far better broadband infrastructure than the US, and it's because of the behavior of our monopoly providers.
I hope this will finally provide an incentive for metropolitan wifi providers and cable vendors to decide upon and implement a reasonable business model. Guys, it isn't rocket science, and it's not going to take anything like the irrational capital-deployment decisions that resulted in the long-haul-fiber overbuild.
One more time: it will take some significant engineering. The old escape towers were just rocket-powered ejection seats with explosive bolts. That technology is not going to get you an intact and fully-functioning crew vehicle that can survive an unpowered fall from 200,000 feet back to an arbitrary spot in the Atlantic Ocean.
IIRC, the Mercury/Gemini/Apollo escape towers were engineered primarily to deal with problems on liftoff. It's not too many minutes into the flight that you're 1000 miles downrange and way out of the breathable atmosphere. What then?
Serious improvements are clearly needed, even if the basic design is workable.
Also, the early towers were very very dangerous- the few times they went off in tests and simulations, they did an awful lot of damage. They were really a last-ditch hope.
I'll buy the analysis which holds the solid-fuel rocket is far more reliable, and of course in a space program where manned launches are "routine," the payoff is huge.
But now you have to engineer a crew-escape mechanism in case of a serious problem during the boost phase. Can we improve on the Mercury-era "escape tower," basically a rocket-powered ejection seat? We need to enable a safe, reliable crew recovery at any point from liftoff till the solid booster burns out, which occurs at a significant altitude and distance downrange.
Notice that the story is about a guy with a partisan interest in the outcome- he's a Thiokol engineer and they make the solid-fuel boosters for the current shuttle. These boosters are the heart of the proposal, and my only question is, do we want solid fuel rockets as the primary lifter for human crews? Don't they present special challenges and risks because they can't be shut down in case of problems? Just asking, IANA astronautical engineer.
Apart from that, this seems like a good blending of proven tech from the shuttle project with a more clear-eyed view of overall project goals, with favorable economics.
The original poster asks, "can't we do better"? As long as we are relying on combustion of chemical fuels, I don't think there's any need to do "better." A quantum leap forward in lift technology will have to await a new type of propulsion, which is at least decades away. In the meantime, let's get busy!
I'm a holder of several software patents, both issued and pending. I vehemently disagree with the whole concept of software patents, but in today's broken environment, you have to apply for these patents as a defensive measure. Otherwise you may find that someone else can prevent you from commercializing your own ideas.
And having been through the process, I can tell you that most examiners at the USPTO are simple, unimpressive bureaucrats with nowhere near the capability and insight needed to evaluate software innovations. They are not operating in bad faith, they just aren't equipped to do the job that the big shots in our industry are asking of them. (Further proof of this point is that they usually retire as soon as they are eligibile, and ever-increasing numbers of them leave the service early. It's a crummy job.)
The path of least resistance for the PTO is to just grant about 60% of the applications put before them and hope that the courts will sort out the conflicts. That's a great solution for the small technology shops and individuals that produce the real innovations, isn't it? Especially since you're now relying on an equally ignorant judge and jury, who will be favorably impressed by those nice, polished, well-dressed attorneys from IBM/Microsoft/Oracle/Toshiba/whoever.
LEGISLATION is needed to solve this problem. The community needs to draft a revision of the patent laws and lobby its passage through Congress. Anybody interested in taking up the challenge?
This was in summer of 1983. I'd been writing Apple and PC programs for years in 6502 and 8088 assembly. I couldn't believe the $10,000 price tag, which was so outlandish you'd never even think of recommending it to clients (a 20meg Corvus hard disk was $3,500 or so at that time, iirc).
But the bitmapped GUI was unlike anything you'd ever imagined. It wasn't even obvious how to start working with the thing. No one else was talking about anything similar at that time, at least not for practical, widespread usage. I think you're wrong about the hundred companies scrambling to implement it. It was one of those ideas whose brilliance took quite a while to become apparent to most people.
About a year later I started hearing rumors out of Redmond about something totally new called "event-driven programming," and of course we know that MSFT dates the original creation of Windows to 1983, with 1.0 shipping in 1985.
It's incredible how much ignorance has been spewed on this thread.
What makes the American system different from all others is our emphasis on freedom. Economic systems derive power directly from the vitality and social cohesion of the societies which generate them. Free societies with high social cohesion generate the most economic value and power because all the incentives align.
American society is losing its social cohesion, mostly because so many people believe all the idiocy being spewed on this thread and everywhere else.
China's government is very successfully motivating huge sectors of its subject population with nationalism and the dream of recovering past greatness. This strategy is profoundly dehumanizing, but it generates tremendous national purpose and cohesion. The last government that tried it was Nazi Germany, and it worked so well there that the rest of the world had to stop them by force.
We like to think that free societies are happy and successful because they are free and open, and in fact the example of their success will encourage others.
China is now trying to prove the opposite. They are trying to control their own people, and motivate them through a shared sense of national purpose and recovery of past greatness.
The last government that tried this was the Nazis. And it took millions of lives to suppress that threat.
The government of China is replaying the experiment. But they have time, numbers, capital, and unlimited reserves of patience on their side.
We are now engaged in the last great test of freedom, people. Wake up, we live in interesting times.
A lot of what is being said about this pair of niche OS's is reminiscent of what was said about Linux back when it was new. A lot of new things get tried, but very very manage to put together that magic combination of meeting a real need with sustained, competent execution.
That means that, as silly as some of you will think, the next big phenomenon comparable to Mosaic or Linux is just getting off the ground now. The few people becoming aware of it now are saying things like "hell, no driver support or PPP? COME ON!"
I'm not saying it's either Syllable or SkyOS but it will certainly be something. And here's the big need that everyone has:
A COMPETENT, FREE DESKTOP O/S.
Windows sucks, OS X is expensive, and Linux is failing on the desktop for a number of reasons, both good and bad. The next big thing will meet this need.
Nah, the late Eighteenth is still the Enlightenment. (Unless you're counting The Sorrows of Young Werther, which would be defensible.) Also, philosophical currents tend to reach music two or three decades after they affect literature and the visual arts.
More directly to your point, think about Schopenhauer's statements on art. That gets you to into the early Nineteenth.
I've never liked that term "Romantic" anyway- too overloaded. To me, Romanticism is primarily about that notion that the human spirit finds deep echoes in the natural world. If you accept that, then the "Romantic" Age doesn't really start until well into the Nineteenth.
Improvisation may be an interesting way of producing pieces of immediate performance art in small forms. And it was considered an essential skill in music pedagogy through the eighteenth century. But it's the wrong way to write larger forms like symphonies. Large software systems are more like symphonies than jazz-sets. They require a lot of long-range planning. The great symphonists of the nineteenth and early twentieth centuries had an incredible gift for long-range thinking, combined with an unbelieveable ability to keep all of a large production in their heads. Here's a possible clue to the required mental equipment: there are anecdotes about most of the the great composers regarding their prodigious, almost inhuman memory-power.
Mozart considered composition a craft. So did Bach, who regularly turned out a new cantata most weeks for his job at the Thomaskirche in Leipzig. The notion that artists have special access to some emotional content not available to ordinary craftsmen is a nineteenth-century idea. But everyone agrees that both Mozart and Bach had access to some pretty unusual stuff- we hear it and respond to it.
The content of programming is perhaps too instrumental (i.e., interesting for its usefulness more than its inherent qualities) to rise to the level of art. But this may be changing with the state-of-the-art games. In a hundred years, people may look back at today's game developers as the inventors of a new art form!
Good point, although 9/11 (more than a year after the bust) was the trigger that destroyed airline equity. Thanks to yield management and the capital-glut, the 1990s was an extremely good decade for the airlines. Remember paying $2500 several times a month for the New York-San Francisco milk run? I sure do ;-)
...and it ain't coming back. What we are seeing is a regression to the mean.
In the Internet days, major enterprises with traditional business models watched in horror as Amazon appeared to make the whole world obsolete. Priceline, nothing more than a travel agent, at one time had a higher market cap than all the major US airlines combined. This was a historic misallocation of capital, comparable to the overinvestment in radio in the 1920s.
Every CEO in America asked himself whether a pimply geek like Jeff Bezos could blow up out of nowhere and destroy his whole business just like that. We know how stupid that was in hindsight, but fear is an incredibly powerful motivator. The result was a tremendous amount of IT overspending. And for those of us who were right in the middle of it, it was rational at the time!
China is finally getting around to doing in software what they have always done in industries deemed "strategic" - they force world-leading companies to share technology with Chinese partners and eventually compete against the foreigners. It works because Western businesses feel like they need exposure to China's purported "billion-person market" at any cost. Look at Boeing. Look at Cisco.
Now it's true that the economic impact in China is negligible because they don't buy software from the West at all - they steal it. And TFA quotes the USTA pathetically whining about how China is closing themselves out of the world's best software. WAKE UP, you fucking morons in the US Dept of Commerce: China wants the "best software in the world" to be made in China, not Palo Alto, not Redmond, not Bangalore, not anywhere else.
This is the first shot in a trade war aimed right at one of the few industries left where the US can honestly be said to have a technological edge.
The Register's take on this (quoted above) indicates that MSFT slashed R&D spending by half, from about $2.5B in last year's Q3 to about $1.5B this year. This can't possibly be right unless their R&D expenditures are not uniform across the year. The $1.4B number is a bit on the low side, you expect a license-based software company to spend between 15% and 18% of revenue on R&D. But $2.5B on $10B in revenue is way the hell out there, just can't be right. Anyone know what the Register's source was?
And BTW, it's striking that Microsoft (or anyone) can spend $5B a year on some of the best programmers in the world, and not get anything new and different to show for it. That's why they have so much cash in the bank: they have nothing to spend it on. That's also why Microsoft's owners (that would be us, the investing public) have been clamoring for the cash to be returned to them in the form of special dividends.
One poster said correctly that, out of 10 deals, a VC looks for 1-2 home runs, maybe 3-4 breakevens and the rest are total losses. What smart VCs do is take a perspective on the market as a whole and bet on what are the coming hot segments. Then they carefully place a few bets in those spaces. To be chosen, a company has to have a top management team, be focused (or re-focusable) on the laser-narrow segment of the market that corresponds to the VC's view, have a "correct" business model, and can check off a whole list of other variables. Everything about the company has to right, down to what color ties the CEO wears (if any).
Basically, a VC manages his risk by only choosing companies that meet a whole range of very narrow constraints, with the only degree of freedom being the specific market segment, and that is chosen by the VC.
This year, the VCs' tea leaves are showing open source as the hot space.
One very interesting comment in TFA was the initial reaction to Fleury's attempts to get funded four years ago: "you must be nuts." Since he didn't fit the VCs' pre-established business-model checkbox at that time, he couldn't get funded. The VC view of the world has changed, and now the "open source" aspect is the hot one.
Another thing good VCs always do is fund to milestones. If you don't hit substantially all of your targets, they will ruthlessly shoot you in the head and not fund your next round.
This will either win big for the early VCs or it will fail. We'll know in about a year (that's a typical length for a funding round).
I looked through a 50,000 line C++ program I wrote last year. There are 83 goto statements- I was damn surprised there were so many! An average of 8 goto's occur in each of ten different functions. Each of the functions is a large state-machine processor (they are actually PDU handlers for network protocols), so in each case the basic function structure is about the same. The goto's are there to break out of the state machine after detecting an error. Each error is completely handled where it occurs but of course invalidates the transaction, which cannot continue. So the goto is actually more natural in this specific case than an exception handler.
If you're going to (judiciously) use goto where appropriate in C++ programs, note one important thing: a goto may not span automatic variable declarations. This should make sense, since the goto target wouldn't how know to reset the stack-frame pointer. This restriction doesn't apply in C, which only allows automatic declarations at the top a function.
There's a presumption built into this whole discussion that the human race is summum valorem, and must be preserved at all costs.
I counterpropose that human beings are more valuable than the human race.
It just might be the height of arrogance to suppose that we should do everything in our considerable power to colonize space simply for the purpose of perpetuating our own spawn, rather than solving the many problems we face here. The proverbial Martian would be well-justified in wanting to prevent a race intent on committing nuclear suicide from spreading over the Universe!
Of course, if we really are alone in the universe, and in fact nothing but a coincidence (in itself an anti-scientific view), then my argument is false.
They counterfeit $100 bills. Fake American hundreds were "legal" tender on Russian black markets for a long while. Fake US currency printed in Lebanon looks better than the real thing, and often appears before the real stuff does too.
Hibernate's fans say it's the "right" way to handle extremely large, complex data models, especially if they already exist.
What you have to do is write yourself a new Java class (according a to big complex set of rules), one for each entity in the model. And there are a lot of different ways to handle various association-types and cascading-persistence requirements.
Once you've done all this, you have Hibernate's methods for "easily" loading and saving the objects.
So in a way you have to re-express an already-existing data model in code.
Anybody with experience in the tool care to argue whether it's worth the effort?
Out of curiosity, how big is your company?
/. for a change :-)
And I really appreciate your dropping the snarky attitude- it's nice to have a civil conversation on
>>>You asked a person apiece at several F500 firms, and think you got an accurate picture of open-source's deployment?
No, I don't think that. Since these people were CIOs or their direct reports, I think I got a picture of how F/OSS deployments are viewed at the top of the food chain. These are the people who are going to shut down all the F/OSS-based skunk-works projects as soon as they start getting important, if they don't like the answers about where support is coming from.