Exactly. Best graduate students go to academia and research labs. Most research money and freedom there. Next tier go to industrial R&D departments. Less money, less freedom. Lowest tier work outside pure R&D in the "business world". As you say, just another programmer where the company doesn't care about their research. Apple is stuck with the bottom because they can't provide the conditions of the first two.
Not true. Which student goes to academia and which one goes to research labs or industrial R&D departments had nothing to do with how good they are. It has more to do with the student's ambition and their understanding of the life.
Not everyone wants to be stuck on a college campus teaching students and doing some obscure research that might not see the light of day. It is just not for everyone. Many bright students want to see their work go into the field, and in the hands of people.
You have a very naive and stupid understanding of people and their motivations.
I've been reading Apple Insider for so many years now so AI means Apple Insider for me. I really had to strain myself to read Artificial Intelligence every time it occurred in the text! Lol..
I hope they make a much much better Siri with neural stuff. Siri might work today, but it's such a poor experience. Today's Siri is like DOS. You need to learn a bunch of commands, and speak them out loud. The only "human factor" of Siri is that she sounds kinda nice.
You're very obviously not a scientist. Go talk to one and ask them about how scientific consensus occurs. People publish a bunch of opinions until it becomes scientifically accepted. That's all. And 25-50 years later when new evidence arises, all of that stuff will be seen as wrong. If you want to know the truth, science will never give it to you.
Example: Newton's laws of motion are wrong. Einstein came along and showed they are broken. 200 years of scientists were mistaken.
Quote:
Originally Posted by Suddenly Newton
Evidence is not the same thing as common sense. You started by defending Bloomberg's use of "common sense." I told you that appealing to common sense is a logical fallacy. Now you talk about evidence? OK, what evidence was presented that Apple's AI efforts were suffering because of secrecy, apart from opinions presented in the article? Evidence is, hard data, measurable facts. It is not opinions. Bloomberg presents none of this, because how could they have any measurable metrics regarding the impact of secrecy on Apple's secret AI efforts? You could say: "but but but Apple published zero papers" as a measure of -- what? That they can't do AI research? Then you have to eliminate other factors such as Apple's desire for secrecy as not being the cause of this metric being zero.
But you're not making that argument. Or anything resembling a logical argument. You posted an ad hom "you're not a scientist go talk to a bunch of scientists" jab at me. Way to dodge addressing my criticisms of your post.
@konqerror, I really want to hear your rebuttal to this. Do you still stick to the "it's obvious" angle, or do you think Bloomberg actually posted evidence that @Suddenly Newton missed?
It is pretty easy to ignore a post when you don't have a genuine response on the Internet. I thought you were better than that.
There is actually a strange mixture of fact and fiction woven through @konqerror's posts in this thread, which puzzles me (as a scientist).
The original point - that research is typically hindered, relatively, by working in isolation from other researchers - is generally true, albeit with caveats. It most notably applies when leading a field (which it is claimed that Apple isn't), since progress is likely to be faster if a larger number of researchers is permitted (via publication, peer review, collaboration and competition) to contribute to the process of advancement. If Apple is doing fundamental AI research in isolation then it quite likely does have some negative impact for those reasons. On the other hand, algorithm, software and technology development in many areas is kept proprietary by commercial entities, trading the risk of a slower pace of progress for competitive advantage in the end.
On the issue of scientists and their motivations with regard to open research, the picture @konqerror presents is inaccurate. There are, indeed, conflicting drivers in planning a scientific career, and in open research fields it is mostly true that publications are key. However, there are rewarding and lucrative opportunities outside those open fields and, in fact, the pay is often much better in commercial and proprietary settings. Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE). Academia is a tough path, at least in any of the good universities, with relatively low starting salaries and intense competition for funding, but does provide the most research freedom and the greatest opportunities for widespread recognition. Companies like Apple and Google offer the possibility of very high wages (but with high demands) in exchange for top talent. So there are lots of choices with different pros and cons, and the result is that there are skilled scientists distributed through all these kinds of organizations.
Most confusing of all are @konqerror's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
I've been reading Apple Insider for so many years now so AI means Apple Insider for me. I really had to strain myself to read Artificial Intelligence every time it occurred in the text! Lol..
I hope they make a much much better Siri with neural stuff. Siri might work today, but it's such a poor experience. Today's Siri is like DOS. You need to learn a bunch of commands, and speak them out loud. The only "human factor" of Siri is that she sounds kinda nice.
Sorry, bullshit. I often mumble, and naturally ask Siri things. She gets it right 95% of the time. I've never felt the need to "speak loudly" or "learn commands", beyond having a rough understanding of what Siri can and can't do, just like any voice assistant. It's not perfect, and never will be (neither will any voice based system) but has made enormous improvements since launched. Please, don't use your odd beliefs as fact.
Sorry, bullshit. I often mumble, and naturally ask Siri things. She gets it right 95% of the time. I've never felt the need to "speak loudly" or "learn commands", beyond having a rough understanding of what Siri can and can't do, just like any voice assistant. It's not perfect, and never will be (neither will any voice based system) but has made enormous improvements since launched. Please, don't use your odd beliefs as fact.
Bullshit? By speak out loud I mean speak instead of type. Sorry mate for shrouding your thoughts. Not like I need to speak louder than normal or anything. The voice recognition is pretty great. The language and meaning, understanding the context, however, is still just an advanced list of commands and actions, isn't it? That's definitely how I feel about it. I don't have the slightest feeling that Siri ever understands me. But rather if I hit the right keywords she'll do the job. Fair enough. But I'd still label it a poor experience, with infinite room for improvement.
Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE).
No. Sure, lower in cash than industry, but much higher than academia. If a DoD lab wants you, they have money. Overall government compensation (GS/S&T) is very competitive after you factor in retirement, job security (impossible to fire), and benefits. On top of that, you have the US Citizen factor. Why do you think the Government is the largest employer in the world? National labs employees are not government employees and don't enjoy the same benefits.
Quote:
Originally Posted by BestKeptSecret
@konqerror, I really want to hear your rebuttal to this. Do you still stick to the "it's obvious" angle, or do you think Bloomberg actually posted evidence that @Suddenly Newton missed?
It is pretty easy to ignore a post when you don't have a genuine response on the Internet. I thought you were better than that.
I'm tired of arguing with people. As a scientist with a solid publication record, what Bloomberg is saying is common sense and I came to the same exact conclusion that Bloomberg did years ago. Apple's secrecy will mean it always sucks in research driven developments because secrecy works against innovation. Not just Apple, defense contractors and people whose research are classified have the same problem, though to less of an extent. If@Suddenly Newton wants to actually get some academic qualifications, then he should.
Quote:
Originally Posted by muppetry
Most confusing of all are @konqerror's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
Science is always a poor approximation based on an incomplete understanding. Worked that way for thousands of years and will continue to do so forever. Science progresses by trying to fix this, but it can never be fixed, so advancement never stops.
You're getting science and engineering confused. Engineering gets by solely on the appropriate approximation and model for the job. This is common, a lot of scientists suck at engineering because they don't realize this and vice versa.
There is actually a strange mixture of fact and fiction woven through <a data-huddler-embed="href" href="/u/188970/konqerror" style="display:inline-block;">@konqerror</a>
's posts in this thread, which puzzles me (as a scientist).
The original point - that research is typically hindered, relatively, by working in isolation from other researchers - is generally true, albeit with caveats. It most notably applies when leading a field (which it is claimed that Apple isn't), since progress is likely to be faster if a larger number of researchers is permitted (via publication, peer review, collaboration and competition) to contribute to the process of advancement. If Apple is doing fundamental AI research in isolation then it quite likely does have some negative impact for those reasons. On the other hand, algorithm, software and technology development in many areas is kept proprietary by commercial entities, trading the risk of a slower pace of progress for competitive advantage in the end.
On the issue of scientists and their motivations with regard to open research, the picture <a data-huddler-embed="href" href="/u/188970/konqerror" style="display:inline-block;">@konqerror</a>
presents is inaccurate. There are, indeed, conflicting drivers in planning a scientific career, and in open research fields it is mostly true that publications are key. However, there are rewarding and lucrative opportunities outside those open fields and, in fact, the pay is often much better in commercial and proprietary settings. Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE). Academia is a tough path, at least in any of the good universities, with relatively low starting salaries and intense competition for funding, but does provide the most research freedom and the greatest opportunities for widespread recognition. Companies like Apple and Google offer the possibility of very high wages (but with high demands) in exchange for top talent. So there are lots of choices with different pros and cons, and the result is that there are skilled scientists distributed through all these kinds of organizations.
Most confusing of all are <a data-huddler-embed="href" href="/u/188970/konqerror" style="display:inline-block;">@konqerror</a>
's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
I find that when someone resorts to attacking (presumptively) their opponent's credentials by saying things like "You're not a scientist" (when he clearly does not know the first thing about me or what my education is) and then starts prattling on about his knowledge of Newton and Einstein, it serves only to undermine his own position and credibility. Konqerror is using a rhetorical argument (ad hominem), which is frequently used in politics, but not in science. As is appealing to "common sense" as an authority, which is nothing more than a specific form of appeal to ignorance, another kind of rhetoric. He talks about science being based on evidence, but then fails to produce the evidence that backs up the argument put forth in Bloomberg's article. As I said, opinion cannot be used as evidence.
I haven't even addressed the points you and another person already have regarding the specifics of konqerror's claims about scientists and science because all of that is irrelevant to my original concerns about the Bloomberg article. But if konqerror is trying to impress me with his grasp of science (and presuming my lack of the same), it is not working. Throwing around a few facts anyone can crib from Wikipedia or YouTube videos does not impress me. And it only serves to pad his verbose dodging of my original post.
I have encountered people IRL who sound like they know what they are talking about, but when you dissect the specifics of what they are actually saying, they are saying nothing: a facade of verbiage. If stated emphatically and confidently, especially in person, it might even fool some people.
He's welcome to respond, but it really isn't necessary at this point. I predict he has nothing of an evidential nature to add at this point. And I would accept nothing else.
Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE).
No. Sure, lower in cash than industry, but much higher than academia. If a DoD lab wants you, they have money. Overall government compensation (GS/S&T) is very competitive after you factor in retirement, job security (impossible to fire), and benefits. On top of that, you have the US Citizen factor. Why do you think the Government is the largest employer in the world? National labs employees are not government employees and don't enjoy the same benefits.
In overall comparison the DoD labs do poorly, certainly compared to the National Labs, where salaries are substantially higher (and always have been), and funding is better. That's not to say that there are not some exceptions on the funding front, but in general the DoD labs are chronically underfunded. Much of the cutting-edge defense research is being done by companies such as Lockheed, Boeing etc., and the National Labs. Benefits at the National Labs are excellent.
Quote:
Quote:Originally Posted by konqerror
Originally Posted by BestKeptSecret
@konqerror, I really want to hear your rebuttal to this. Do you still stick to the "it's obvious" angle, or do you think Bloomberg actually posted evidence that @Suddenly Newton missed?
It is pretty easy to ignore a post when you don't have a genuine response on the Internet. I thought you were better than that.
I'm tired of arguing with people. As a scientist with a solid publication record, what Bloomberg is saying is common sense and I came to the same exact conclusion that Bloomberg did years ago. Apple's secrecy will mean it always sucks in research driven developments because secrecy works against innovation. Not just Apple, defense contractors and people whose research are classified have the same problem, though to less of an extent. If@Suddenly Newton wants to actually get some academic qualifications, then he should.
Again - this is problematic. You really are merely asserting, and that makes me wary. Have you ever been involved in classified or commercially proprietary research, or are you just guessing here? There are, as I mentioned previously, legitimate competing drivers when it comes to openness in research and, at best, you are grossly oversimplifying the picture.
Quote:
Quote:Originally Posted by konqerror
Originally Posted by muppetry
Most confusing of all are @konqerror's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
Science is always a poor approximation based on an incomplete understanding. Worked that way for thousands of years and will continue to do so forever. Science progresses by trying to fix this, but it can never be fixed, so advancement never stops.
You're getting science and engineering confused. Engineering gets by solely on the appropriate approximation and model for the job. This is common, a lot of scientists suck at engineering because they don't realize this and vice versa.
I get the sense that you are just rambling now. Previously it was "science will never give you the truth", now it is "science is always a poor approximation based on an incomplete understanding". These are meaningless slogans, not intelligent observations, and the confusion seems to be all yours. What are you trying to argue - that Newtonian mechanics was not real science, but just an engineering approximation? Frankly, after reading your posts, I do not find your claimed scientific credentials to be even remotely credible.
There is actually a strange mixture of fact and fiction woven through @konqerror
's posts in this thread, which puzzles me (as a scientist).
The original point - that research is typically hindered, relatively, by working in isolation from other researchers - is generally true, albeit with caveats. It most notably applies when leading a field (which it is claimed that Apple isn't), since progress is likely to be faster if a larger number of researchers is permitted (via publication, peer review, collaboration and competition) to contribute to the process of advancement. If Apple is doing fundamental AI research in isolation then it quite likely does have some negative impact for those reasons. On the other hand, algorithm, software and technology development in many areas is kept proprietary by commercial entities, trading the risk of a slower pace of progress for competitive advantage in the end.
On the issue of scientists and their motivations with regard to open research, the picture @konqerror
presents is inaccurate. There are, indeed, conflicting drivers in planning a scientific career, and in open research fields it is mostly true that publications are key. However, there are rewarding and lucrative opportunities outside those open fields and, in fact, the pay is often much better in commercial and proprietary settings. Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE). Academia is a tough path, at least in any of the good universities, with relatively low starting salaries and intense competition for funding, but does provide the most research freedom and the greatest opportunities for widespread recognition. Companies like Apple and Google offer the possibility of very high wages (but with high demands) in exchange for top talent. So there are lots of choices with different pros and cons, and the result is that there are skilled scientists distributed through all these kinds of organizations.
's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
I find that when someone resorts to attacking (presumptively) their opponent's credentials by saying things like "You're not a scientist" (when he clearly does not know the first thing about me or what my education is) and then starts prattling on about his knowledge of Newton and Einstein, it serves only to undermine his own position and credibility. Konqerror is using a rhetorical argument (ad hominem), which is frequently used in politics, but not in science. As is appealing to "common sense" as an authority, which is nothing more than a specific form of appeal to ignorance, another kind of rhetoric. He talks about science being based on evidence, but then fails to produce the evidence that backs up the argument put forth in Bloomberg's article. As I said, opinion cannot be used as evidence.
I haven't even addressed the points you and another person already have regarding the specifics of konqerror's claims about scientists and science because all of that is irrelevant to my original concerns about the Bloomberg article. But if konqerror is trying to impress me with his grasp of science (and presuming my lack of the same), it is not working. Throwing around a few facts anyone can crib from Wikipedia or YouTube videos does not impress me. And it only serves to pad his verbose dodging of my original post.
I have encountered people IRL who sound like they know what they are talking about, but when you dissect the specifics of what they are actually saying, they are saying nothing: a facade of verbiage. If stated emphatically and confidently, especially in person, it might even fool some people.
He's welcome to respond, but it really isn't necessary at this point. I predict he has nothing of an evidential nature to add at this point. And I would accept nothing else.
I have to agree with your assessment that his case entirely lacks specifics and hinges on an appeal to authority - his own. He sounds closer to the real thing than many "fake" scientists who post, but then spoils it with the ad homs, unscientific argument and apparent ignorance of the realities of research. If he really is in science then he's a weird one.
In overall comparison the DoD labs do poorly, certainly compared to the National Labs, where salaries are substantially higher (and always have been), and funding is better. That's not to say that there are not some exceptions on the funding front, but in general the DoD labs are chronically underfunded. Much of the cutting-edge defense research is being done by companies such as Lockheed, Boeing etc., and the National Labs. Benefits at the National Labs are excellent.
You're not a researcher, if you do, you're not involved in getting money. DoD labs do 6.1/basic research. Yes, their quality varies, out of the majors, one is great, one is OK, one sucks. Lockheed and Boeing do very little 6.1. In fact, Boeing's 6.1 operation is nearly independent, it's called HRL and it does receive significantly less money than you'd expect. National labs are a mix but they are definitely in the 6.1 game.
You can't compare 6.1 research to building a missile (6.3 and above). Totally different funding sources and totally different amounts of money.
Quote:
Originally Posted by muppetry
I have to agree with your assessment that his case entirely lacks specifics and hinges on an appeal to authority - his own. He sounds closer to the real thing than many "fake" scientists who post, but then spoils it with the ad homs, unscientific argument and apparent ignorance of the realities of research. If he really is in science then he's a weird one.
You have a masters or are a grad student. I can tell from the above argument.
In overall comparison the DoD labs do poorly, certainly compared to the National Labs, where salaries are substantially higher (and always have been), and funding is better. That's not to say that there are not some exceptions on the funding front, but in general the DoD labs are chronically underfunded. Much of the cutting-edge defense research is being done by companies such as Lockheed, Boeing etc., and the National Labs. Benefits at the National Labs are excellent.
You're not a researcher, if you do, you're not involved in getting money. DoD labs do 6.1/basic research. Yes, their quality varies, out of the majors, one is great, one is OK, one sucks. Lockheed and Boeing do very little 6.1. In fact, Boeing's 6.1 operation is nearly independent, it's called HRL and it does receive significantly less money than you'd expect. National labs are a mix but they are definitely in the 6.1 game.
You can't compare 6.1 research to building a missile (6.3 and above). Totally different funding sources and totally different amounts of money.
Quote:
Originally Posted by muppetry
I have to agree with your assessment that his case entirely lacks specifics and hinges on an appeal to authority - his own. He sounds closer to the real thing than many "fake" scientists who post, but then spoils it with the ad homs, unscientific argument and apparent ignorance of the realities of research. If he really is in science then he's a weird one.
You have a masters or are a grad student. I can tell from the above argument.
Now I'm actually intrigued. Your posts are such a bizarre mix of factual content (though mostly off-topic) and strange misconceptions. I won't bother to argue with your assessment of the labs, except to point out that it does not support (because it is unrelated to) your premise (that closed research sucks), or counter my point (that the DoD labs are not generally well-funded or particularly well-paid), or address in any way the issue of closed versus open research.
I am surprised that you went so far out on a limb to try to guess the details of my background, especially since, remarkably, you managed to get every single one wrong. If you re-apply your deductive powers to that statement and reread my posts then even mediocre reasoning should get you much closer.
Anyway - as @Suddenly Newton said previously - if you have anything substantive to contribute to the discussion then feel free to do so, but if you just want to make pejorative comments about other peoples' qualifications or status in an attempt to discredit their arguments then there's not much to talk about. Just as a reminder - the topic at hand was whether lack of openness was hurting Apple in the field of AI, and the question was whether you had any insights beyond "it obviously is".
You're very obviously not a scientist. Go talk to one and ask them about how scientific consensus occurs. People publish a bunch of opinions until it becomes scientifically accepted. That's all. And 25-50 years later when new evidence arises, all of that stuff will be seen as wrong. If you want to know the truth, science will never give it to you.
Example: Newton's laws of motion are wrong. Einstein came along and showed they are broken. 200 years of scientists were mistaken.
Scientists can make conclusions based on only the information they have available. If new information comes to light that contradicts the previous conclusion, there's nothing wrong with that.
And you're confusing Newton's laws of motion with Newtonian mechanics. Einstein simply showed that Newtonian mechanics was inadequate at very small scales and very large scales such as in space. Newton observed certain physical phenomena and developed theories based on his observations with the information that he had available. Einstein found, however, that Newtonian mechanics could not explain phenomena at very large or very small scales. But Newtonian mechanics works at most length scales that we see in everyday life.
Honesty, this is nothing new for Apple, it was not unusally for Apple employees to got to trade shows and conference registered as working for another company to fake companies. The reason this is an issue is the fact that most people in AI were stuck in some university for all these years since no one would hire them since they did not really know what to do with it. They are so use to taking other peoples ideas and running with them or sharing their ideas since they could not figure out what to do next. Now they being told not to share anymore and they do not know how to function.
You sound like a really spiteful person.
Industry has different requirements. They might have very specific goals related to an intended product line. Research is often applicable to more than one thing. Others extend it, because they can use the results for something they may have in mind that didn't seem previously feasible.
As others have mentioned, researchers may also be concerned about their own careers. They might leave Apple one day, and without any public record of their work, they may no longer be employable at the same level.
Exactly. Best graduate students go to academia and research labs. Most research money and freedom there. Next tier go to industrial R&D departments. Less money, less freedom. Lowest tier work outside pure R&D in the "business world". As you say, just another programmer where the company doesn't care about their research. Apple is stuck with the bottom because they can't provide the conditions of the first two.
That's a specious statement that the "best" students stay in academia. Everyone has their personal strengths and preferences. Academia receives relatively little development (product-focused) funding. In EE/CS the most R&D money and greatest potential for monetary reward is in industry--and that's where most of the real-world talent goes. Basic research may be more likely to occur in academic and research labs, but the part that has practical applications is few and far between. Most of the research is just part of training new employees for the private sector.
Remember the cogent adage: Those who can, do, and those who can't, teach.
Honesty, this is nothing new for Apple, it was not unusally for Apple employees to got to trade shows and conference registered as working for another company to fake companies. The reason this is an issue is the fact that most people in AI were stuck in some university for all these years since no one would hire them since they did not really know what to do with it. They are so use to taking other peoples ideas and running with them or sharing their ideas since they could not figure out what to do next. Now they being told not to share anymore and they do not know how to function.
You sound like a really spiteful person.
Industry has different requirements. They might have very specific goals related to an intended product line. Research is often applicable to more than one thing. Others extend it, because they can use the results for something they may have in mind that didn't seem previously feasible.
As others have mentioned, researchers may also be concerned about their own careers. They might leave Apple one day, and without any public record of their work, they may no longer be employable at the same level.
I totally agree. If a university professor publishes a seminal paper or writes a well-received book, the professor's name remains associated with that work forever. But in a corporation, the scientist who developed a critical piece of technology will be forgotten. People will remember the CEOs and the sales/marketing people.
There is so much wrong with this post I don't know where to start. To think that people in AI are "stuck" in some university because no one would hire them is one of the most asinine ideas I've ever heard. On the contrary, for those who are truly passionate about doing AI research, there is no better place to be than a university because that's where the real, cutting-edge research takes place. Many of the basic technologies that make things like the iPhone possible were invented by some of those very people who you think are "stuck" in a university because they couldn't go to industry. Where do you think Siri came from? It was a DARPA project undertaken in a joint effort involving dozens of computer scientists at the leading American universities. Do you really think people on that project were "stuck" as a university professor because no company would hire them?
And you criticize them for "taking other people's ideas and running with them" or "sharing their ideas since they could not figure out what to do next." Where do you think the mouse came from? It came from the very practice that you are deriding. The free exchange of ideas is critical to driving the progress of research. Science will not progress with people curled up in a corner. Going to conferences and publishing papers is vital to professors getting their name known in the community. Those conferences help them get their next grant. It helps professors' graduate students get placed.
Where did you think the money came from to fund those ideas. Usually the government you mention DARPA, which happen to provide the start up funds for a company I worked for back in the 90's which invented ATM networking. The founders were four Professors with an idea but could not take it to the next level IE a product until DARPA pointed them in the right direction an through lots of money at them. When in College I had a few professors who whole jobs was to get funding for ideas as well as going to companies and asking them to let the school work on idea for them. Many times companies will go to Universities, through money their way to solve a problem, some time it is unique solution but other time anyone could have come up with the idea.
Those individuals real issue is the fact Apple is not throwing money at Universities or its professors to have them advance an idea or develop something new to solve a unique problem. Here is the down side of this, when a company pays a University to come up with an idea the Company does not own the solution or the IP the University and the Professor and those students who worked on the idea own it along with the company. Do you really think Apple wants to share their ideas with a University. Where do you think all these patent trolls come from these are companys who buy IP from Universities who can not turn a profit from the idea and they sell them to get some money out of what they have laying around.
Yeah I am not interest in hearing a bunch people who spent their entire life doing one things complaining that getting paid big $ to keep their month shut is a bad thing.
Trust me I get the whole idea about sharing ideas to advance an idea, thus the whole idea of an institution. But when someone paying your bills they get to decide how you will will behave. Most of these people want the money but not the strings which are attached.
Comments
Exactly. Best graduate students go to academia and research labs. Most research money and freedom there. Next tier go to industrial R&D departments. Less money, less freedom. Lowest tier work outside pure R&D in the "business world". As you say, just another programmer where the company doesn't care about their research. Apple is stuck with the bottom because they can't provide the conditions of the first two.
Not true. Which student goes to academia and which one goes to research labs or industrial R&D departments had nothing to do with how good they are. It has more to do with the student's ambition and their understanding of the life.
Not everyone wants to be stuck on a college campus teaching students and doing some obscure research that might not see the light of day. It is just not for everyone. Many bright students want to see their work go into the field, and in the hands of people.
You have a very naive and stupid understanding of people and their motivations.
I hope they make a much much better Siri with neural stuff. Siri might work today, but it's such a poor experience. Today's Siri is like DOS. You need to learn a bunch of commands, and speak them out loud. The only "human factor" of Siri is that she sounds kinda nice.
You're very obviously not a scientist. Go talk to one and ask them about how scientific consensus occurs. People publish a bunch of opinions until it becomes scientifically accepted. That's all. And 25-50 years later when new evidence arises, all of that stuff will be seen as wrong. If you want to know the truth, science will never give it to you.
Example: Newton's laws of motion are wrong. Einstein came along and showed they are broken. 200 years of scientists were mistaken.
Evidence is not the same thing as common sense. You started by defending Bloomberg's use of "common sense." I told you that appealing to common sense is a logical fallacy. Now you talk about evidence? OK, what evidence was presented that Apple's AI efforts were suffering because of secrecy, apart from opinions presented in the article? Evidence is, hard data, measurable facts. It is not opinions. Bloomberg presents none of this, because how could they have any measurable metrics regarding the impact of secrecy on Apple's secret AI efforts? You could say: "but but but Apple published zero papers" as a measure of -- what? That they can't do AI research? Then you have to eliminate other factors such as Apple's desire for secrecy as not being the cause of this metric being zero.
But you're not making that argument. Or anything resembling a logical argument. You posted an ad hom "you're not a scientist go talk to a bunch of scientists" jab at me. Way to dodge addressing my criticisms of your post.
@konqerror, I really want to hear your rebuttal to this. Do you still stick to the "it's obvious" angle, or do you think Bloomberg actually posted evidence that @Suddenly Newton missed?
It is pretty easy to ignore a post when you don't have a genuine response on the Internet. I thought you were better than that.
There is actually a strange mixture of fact and fiction woven through @konqerror's posts in this thread, which puzzles me (as a scientist).
The original point - that research is typically hindered, relatively, by working in isolation from other researchers - is generally true, albeit with caveats. It most notably applies when leading a field (which it is claimed that Apple isn't), since progress is likely to be faster if a larger number of researchers is permitted (via publication, peer review, collaboration and competition) to contribute to the process of advancement. If Apple is doing fundamental AI research in isolation then it quite likely does have some negative impact for those reasons. On the other hand, algorithm, software and technology development in many areas is kept proprietary by commercial entities, trading the risk of a slower pace of progress for competitive advantage in the end.
On the issue of scientists and their motivations with regard to open research, the picture @konqerror presents is inaccurate. There are, indeed, conflicting drivers in planning a scientific career, and in open research fields it is mostly true that publications are key. However, there are rewarding and lucrative opportunities outside those open fields and, in fact, the pay is often much better in commercial and proprietary settings. Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE). Academia is a tough path, at least in any of the good universities, with relatively low starting salaries and intense competition for funding, but does provide the most research freedom and the greatest opportunities for widespread recognition. Companies like Apple and Google offer the possibility of very high wages (but with high demands) in exchange for top talent. So there are lots of choices with different pros and cons, and the result is that there are skilled scientists distributed through all these kinds of organizations.
Most confusing of all are @konqerror's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
Yep.
Sorry, bullshit. I often mumble, and naturally ask Siri things. She gets it right 95% of the time. I've never felt the need to "speak loudly" or "learn commands", beyond having a rough understanding of what Siri can and can't do, just like any voice assistant. It's not perfect, and never will be (neither will any voice based system) but has made enormous improvements since launched. Please, don't use your odd beliefs as fact.
Experiences never vary, it's a known fact.
Well, that’s an outright lie.
Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE).
No. Sure, lower in cash than industry, but much higher than academia. If a DoD lab wants you, they have money. Overall government compensation (GS/S&T) is very competitive after you factor in retirement, job security (impossible to fire), and benefits. On top of that, you have the US Citizen factor. Why do you think the Government is the largest employer in the world? National labs employees are not government employees and don't enjoy the same benefits.
@konqerror, I really want to hear your rebuttal to this. Do you still stick to the "it's obvious" angle, or do you think Bloomberg actually posted evidence that @Suddenly Newton missed?
It is pretty easy to ignore a post when you don't have a genuine response on the Internet. I thought you were better than that.
I'm tired of arguing with people. As a scientist with a solid publication record, what Bloomberg is saying is common sense and I came to the same exact conclusion that Bloomberg did years ago. Apple's secrecy will mean it always sucks in research driven developments because secrecy works against innovation. Not just Apple, defense contractors and people whose research are classified have the same problem, though to less of an extent. If @Suddenly Newton wants to actually get some academic qualifications, then he should.
Most confusing of all are @konqerror's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
Science is always a poor approximation based on an incomplete understanding. Worked that way for thousands of years and will continue to do so forever. Science progresses by trying to fix this, but it can never be fixed, so advancement never stops.
You're getting science and engineering confused. Engineering gets by solely on the appropriate approximation and model for the job. This is common, a lot of scientists suck at engineering because they don't realize this and vice versa.
I find that when someone resorts to attacking (presumptively) their opponent's credentials by saying things like "You're not a scientist" (when he clearly does not know the first thing about me or what my education is) and then starts prattling on about his knowledge of Newton and Einstein, it serves only to undermine his own position and credibility. Konqerror is using a rhetorical argument (ad hominem), which is frequently used in politics, but not in science. As is appealing to "common sense" as an authority, which is nothing more than a specific form of appeal to ignorance, another kind of rhetoric. He talks about science being based on evidence, but then fails to produce the evidence that backs up the argument put forth in Bloomberg's article. As I said, opinion cannot be used as evidence.
I haven't even addressed the points you and another person already have regarding the specifics of konqerror's claims about scientists and science because all of that is irrelevant to my original concerns about the Bloomberg article. But if konqerror is trying to impress me with his grasp of science (and presuming my lack of the same), it is not working. Throwing around a few facts anyone can crib from Wikipedia or YouTube videos does not impress me. And it only serves to pad his verbose dodging of my original post.
I have encountered people IRL who sound like they know what they are talking about, but when you dissect the specifics of what they are actually saying, they are saying nothing: a facade of verbiage. If stated emphatically and confidently, especially in person, it might even fool some people.
He's welcome to respond, but it really isn't necessary at this point. I predict he has nothing of an evidential nature to add at this point. And I would accept nothing else.
Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE).
No. Sure, lower in cash than industry, but much higher than academia. If a DoD lab wants you, they have money. Overall government compensation (GS/S&T) is very competitive after you factor in retirement, job security (impossible to fire), and benefits. On top of that, you have the US Citizen factor. Why do you think the Government is the largest employer in the world? National labs employees are not government employees and don't enjoy the same benefits.
In overall comparison the DoD labs do poorly, certainly compared to the National Labs, where salaries are substantially higher (and always have been), and funding is better. That's not to say that there are not some exceptions on the funding front, but in general the DoD labs are chronically underfunded. Much of the cutting-edge defense research is being done by companies such as Lockheed, Boeing etc., and the National Labs. Benefits at the National Labs are excellent.
Quote:Originally Posted by konqerror
@konqerror, I really want to hear your rebuttal to this. Do you still stick to the "it's obvious" angle, or do you think Bloomberg actually posted evidence that @Suddenly Newton missed?
It is pretty easy to ignore a post when you don't have a genuine response on the Internet. I thought you were better than that.
I'm tired of arguing with people. As a scientist with a solid publication record, what Bloomberg is saying is common sense and I came to the same exact conclusion that Bloomberg did years ago. Apple's secrecy will mean it always sucks in research driven developments because secrecy works against innovation. Not just Apple, defense contractors and people whose research are classified have the same problem, though to less of an extent. If @Suddenly Newton wants to actually get some academic qualifications, then he should.
Again - this is problematic. You really are merely asserting, and that makes me wary. Have you ever been involved in classified or commercially proprietary research, or are you just guessing here? There are, as I mentioned previously, legitimate competing drivers when it comes to openness in research and, at best, you are grossly oversimplifying the picture.
Quote:Originally Posted by konqerror
Most confusing of all are @konqerror's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
Science is always a poor approximation based on an incomplete understanding. Worked that way for thousands of years and will continue to do so forever. Science progresses by trying to fix this, but it can never be fixed, so advancement never stops.
You're getting science and engineering confused. Engineering gets by solely on the appropriate approximation and model for the job. This is common, a lot of scientists suck at engineering because they don't realize this and vice versa.
I get the sense that you are just rambling now. Previously it was "science will never give you the truth", now it is "science is always a poor approximation based on an incomplete understanding". These are meaningless slogans, not intelligent observations, and the confusion seems to be all yours. What are you trying to argue - that Newtonian mechanics was not real science, but just an engineering approximation? Frankly, after reading your posts, I do not find your claimed scientific credentials to be even remotely credible.
There is actually a strange mixture of fact and fiction woven through @konqerror
's posts in this thread, which puzzles me (as a scientist).
The original point - that research is typically hindered, relatively, by working in isolation from other researchers - is generally true, albeit with caveats. It most notably applies when leading a field (which it is claimed that Apple isn't), since progress is likely to be faster if a larger number of researchers is permitted (via publication, peer review, collaboration and competition) to contribute to the process of advancement. If Apple is doing fundamental AI research in isolation then it quite likely does have some negative impact for those reasons. On the other hand, algorithm, software and technology development in many areas is kept proprietary by commercial entities, trading the risk of a slower pace of progress for competitive advantage in the end.
On the issue of scientists and their motivations with regard to open research, the picture @konqerror
presents is inaccurate. There are, indeed, conflicting drivers in planning a scientific career, and in open research fields it is mostly true that publications are key. However, there are rewarding and lucrative opportunities outside those open fields and, in fact, the pay is often much better in commercial and proprietary settings. Contrary to his comments on his personal experience, DoD salaries are not that attractive compared to similar opportunities in industry, or even to the National Labs (DoE). Academia is a tough path, at least in any of the good universities, with relatively low starting salaries and intense competition for funding, but does provide the most research freedom and the greatest opportunities for widespread recognition. Companies like Apple and Google offer the possibility of very high wages (but with high demands) in exchange for top talent. So there are lots of choices with different pros and cons, and the result is that there are skilled scientists distributed through all these kinds of organizations.
Most confusing of all are @konqerror
's curious comments about science not providing the truth and always being wrong, which seem to imply a fundamental misunderstanding of how science works. Even if one started with the premise that Newtonian mechanics were "wrong" it is a logical fallacy to use one instance to argue that therefore science is always going to be wrong. But, in fact, Newtonian mechanics is not wrong, it is simply an approximation that is entirely adequate to describe almost all physical systems that we interact with. And just because Special Relativity addresses deficiencies in classical EM does not mean that it is a prerequisite for understanding, using, or even designing an electric motor or cell phone, which are demonstrably non-relativistic devices.
I find that when someone resorts to attacking (presumptively) their opponent's credentials by saying things like "You're not a scientist" (when he clearly does not know the first thing about me or what my education is) and then starts prattling on about his knowledge of Newton and Einstein, it serves only to undermine his own position and credibility. Konqerror is using a rhetorical argument (ad hominem), which is frequently used in politics, but not in science. As is appealing to "common sense" as an authority, which is nothing more than a specific form of appeal to ignorance, another kind of rhetoric. He talks about science being based on evidence, but then fails to produce the evidence that backs up the argument put forth in Bloomberg's article. As I said, opinion cannot be used as evidence.
I haven't even addressed the points you and another person already have regarding the specifics of konqerror's claims about scientists and science because all of that is irrelevant to my original concerns about the Bloomberg article. But if konqerror is trying to impress me with his grasp of science (and presuming my lack of the same), it is not working. Throwing around a few facts anyone can crib from Wikipedia or YouTube videos does not impress me. And it only serves to pad his verbose dodging of my original post.
I have encountered people IRL who sound like they know what they are talking about, but when you dissect the specifics of what they are actually saying, they are saying nothing: a facade of verbiage. If stated emphatically and confidently, especially in person, it might even fool some people.
He's welcome to respond, but it really isn't necessary at this point. I predict he has nothing of an evidential nature to add at this point. And I would accept nothing else.
I have to agree with your assessment that his case entirely lacks specifics and hinges on an appeal to authority - his own. He sounds closer to the real thing than many "fake" scientists who post, but then spoils it with the ad homs, unscientific argument and apparent ignorance of the realities of research. If he really is in science then he's a weird one.
In overall comparison the DoD labs do poorly, certainly compared to the National Labs, where salaries are substantially higher (and always have been), and funding is better. That's not to say that there are not some exceptions on the funding front, but in general the DoD labs are chronically underfunded. Much of the cutting-edge defense research is being done by companies such as Lockheed, Boeing etc., and the National Labs. Benefits at the National Labs are excellent.
You're not a researcher, if you do, you're not involved in getting money. DoD labs do 6.1/basic research. Yes, their quality varies, out of the majors, one is great, one is OK, one sucks. Lockheed and Boeing do very little 6.1. In fact, Boeing's 6.1 operation is nearly independent, it's called HRL and it does receive significantly less money than you'd expect. National labs are a mix but they are definitely in the 6.1 game.
You can't compare 6.1 research to building a missile (6.3 and above). Totally different funding sources and totally different amounts of money.
I have to agree with your assessment that his case entirely lacks specifics and hinges on an appeal to authority - his own. He sounds closer to the real thing than many "fake" scientists who post, but then spoils it with the ad homs, unscientific argument and apparent ignorance of the realities of research. If he really is in science then he's a weird one.
You have a masters or are a grad student. I can tell from the above argument.
In overall comparison the DoD labs do poorly, certainly compared to the National Labs, where salaries are substantially higher (and always have been), and funding is better. That's not to say that there are not some exceptions on the funding front, but in general the DoD labs are chronically underfunded. Much of the cutting-edge defense research is being done by companies such as Lockheed, Boeing etc., and the National Labs. Benefits at the National Labs are excellent.
You're not a researcher, if you do, you're not involved in getting money. DoD labs do 6.1/basic research. Yes, their quality varies, out of the majors, one is great, one is OK, one sucks. Lockheed and Boeing do very little 6.1. In fact, Boeing's 6.1 operation is nearly independent, it's called HRL and it does receive significantly less money than you'd expect. National labs are a mix but they are definitely in the 6.1 game.
You can't compare 6.1 research to building a missile (6.3 and above). Totally different funding sources and totally different amounts of money.
I have to agree with your assessment that his case entirely lacks specifics and hinges on an appeal to authority - his own. He sounds closer to the real thing than many "fake" scientists who post, but then spoils it with the ad homs, unscientific argument and apparent ignorance of the realities of research. If he really is in science then he's a weird one.
You have a masters or are a grad student. I can tell from the above argument.
Now I'm actually intrigued. Your posts are such a bizarre mix of factual content (though mostly off-topic) and strange misconceptions. I won't bother to argue with your assessment of the labs, except to point out that it does not support (because it is unrelated to) your premise (that closed research sucks), or counter my point (that the DoD labs are not generally well-funded or particularly well-paid), or address in any way the issue of closed versus open research.
I am surprised that you went so far out on a limb to try to guess the details of my background, especially since, remarkably, you managed to get every single one wrong. If you re-apply your deductive powers to that statement and reread my posts then even mediocre reasoning should get you much closer.
Anyway - as @Suddenly Newton said previously - if you have anything substantive to contribute to the discussion then feel free to do so, but if you just want to make pejorative comments about other peoples' qualifications or status in an attempt to discredit their arguments then there's not much to talk about. Just as a reminder - the topic at hand was whether lack of openness was hurting Apple in the field of AI, and the question was whether you had any insights beyond "it obviously is".
You're very obviously not a scientist. Go talk to one and ask them about how scientific consensus occurs. People publish a bunch of opinions until it becomes scientifically accepted. That's all. And 25-50 years later when new evidence arises, all of that stuff will be seen as wrong. If you want to know the truth, science will never give it to you.
Example: Newton's laws of motion are wrong. Einstein came along and showed they are broken. 200 years of scientists were mistaken.
Scientists can make conclusions based on only the information they have available. If new information comes to light that contradicts the previous conclusion, there's nothing wrong with that.
And you're confusing Newton's laws of motion with Newtonian mechanics. Einstein simply showed that Newtonian mechanics was inadequate at very small scales and very large scales such as in space. Newton observed certain physical phenomena and developed theories based on his observations with the information that he had available. Einstein found, however, that Newtonian mechanics could not explain phenomena at very large or very small scales. But Newtonian mechanics works at most length scales that we see in everyday life.
Honesty, this is nothing new for Apple, it was not unusally for Apple employees to got to trade shows and conference registered as working for another company to fake companies. The reason this is an issue is the fact that most people in AI were stuck in some university for all these years since no one would hire them since they did not really know what to do with it. They are so use to taking other peoples ideas and running with them or sharing their ideas since they could not figure out what to do next. Now they being told not to share anymore and they do not know how to function.
You sound like a really spiteful person.
Industry has different requirements. They might have very specific goals related to an intended product line. Research is often applicable to more than one thing. Others extend it, because they can use the results for something they may have in mind that didn't seem previously feasible.
As others have mentioned, researchers may also be concerned about their own careers. They might leave Apple one day, and without any public record of their work, they may no longer be employable at the same level.
Exactly. Best graduate students go to academia and research labs. Most research money and freedom there. Next tier go to industrial R&D departments. Less money, less freedom. Lowest tier work outside pure R&D in the "business world". As you say, just another programmer where the company doesn't care about their research. Apple is stuck with the bottom because they can't provide the conditions of the first two.
That's a specious statement that the "best" students stay in academia. Everyone has their personal strengths and preferences. Academia receives relatively little development (product-focused) funding. In EE/CS the most R&D money and greatest potential for monetary reward is in industry--and that's where most of the real-world talent goes. Basic research may be more likely to occur in academic and research labs, but the part that has practical applications is few and far between. Most of the research is just part of training new employees for the private sector.
Remember the cogent adage: Those who can, do, and those who can't, teach.
Honesty, this is nothing new for Apple, it was not unusally for Apple employees to got to trade shows and conference registered as working for another company to fake companies. The reason this is an issue is the fact that most people in AI were stuck in some university for all these years since no one would hire them since they did not really know what to do with it. They are so use to taking other peoples ideas and running with them or sharing their ideas since they could not figure out what to do next. Now they being told not to share anymore and they do not know how to function.
You sound like a really spiteful person.
Industry has different requirements. They might have very specific goals related to an intended product line. Research is often applicable to more than one thing. Others extend it, because they can use the results for something they may have in mind that didn't seem previously feasible.
As others have mentioned, researchers may also be concerned about their own careers. They might leave Apple one day, and without any public record of their work, they may no longer be employable at the same level.
I totally agree. If a university professor publishes a seminal paper or writes a well-received book, the professor's name remains associated with that work forever. But in a corporation, the scientist who developed a critical piece of technology will be forgotten. People will remember the CEOs and the sales/marketing people.
There is so much wrong with this post I don't know where to start. To think that people in AI are "stuck" in some university because no one would hire them is one of the most asinine ideas I've ever heard. On the contrary, for those who are truly passionate about doing AI research, there is no better place to be than a university because that's where the real, cutting-edge research takes place. Many of the basic technologies that make things like the iPhone possible were invented by some of those very people who you think are "stuck" in a university because they couldn't go to industry. Where do you think Siri came from? It was a DARPA project undertaken in a joint effort involving dozens of computer scientists at the leading American universities. Do you really think people on that project were "stuck" as a university professor because no company would hire them?
And you criticize them for "taking other people's ideas and running with them" or "sharing their ideas since they could not figure out what to do next." Where do you think the mouse came from? It came from the very practice that you are deriding. The free exchange of ideas is critical to driving the progress of research. Science will not progress with people curled up in a corner. Going to conferences and publishing papers is vital to professors getting their name known in the community. Those conferences help them get their next grant. It helps professors' graduate students get placed.
Where did you think the money came from to fund those ideas. Usually the government you mention DARPA, which happen to provide the start up funds for a company I worked for back in the 90's which invented ATM networking. The founders were four Professors with an idea but could not take it to the next level IE a product until DARPA pointed them in the right direction an through lots of money at them. When in College I had a few professors who whole jobs was to get funding for ideas as well as going to companies and asking them to let the school work on idea for them. Many times companies will go to Universities, through money their way to solve a problem, some time it is unique solution but other time anyone could have come up with the idea.
Those individuals real issue is the fact Apple is not throwing money at Universities or its professors to have them advance an idea or develop something new to solve a unique problem. Here is the down side of this, when a company pays a University to come up with an idea the Company does not own the solution or the IP the University and the Professor and those students who worked on the idea own it along with the company. Do you really think Apple wants to share their ideas with a University. Where do you think all these patent trolls come from these are companys who buy IP from Universities who can not turn a profit from the idea and they sell them to get some money out of what they have laying around.
Yeah I am not interest in hearing a bunch people who spent their entire life doing one things complaining that getting paid big $ to keep their month shut is a bad thing.
Trust me I get the whole idea about sharing ideas to advance an idea, thus the whole idea of an institution. But when someone paying your bills they get to decide how you will will behave. Most of these people want the money but not the strings which are attached.