I would put it at 'crap shoot'. Journals and patents have the same basic problem of 'depends on who you get', with wildly inconsistent standards, and in both cases they become trivial to handle when you hire on people who's domain is dealing with patents/publications. Same with research grants.
The military always has lots of research going on that is not public, but while it is often technically impressive to people in the relevant fields, it is almost always pretty mundane sounding compared to public imagination.
I would not be surprised if they had some new launch vehicles and better communications/observation equipment than the civilian sector, I would not even be surprised if some department had something like a functional linear aerospike based launch system or some other incremental improvement like that.
But in general, people working in research tend to have a good idea of what the current possibility space is, and secret military projects that come out later are rarely more than 'oh wow,they got that thing we all knew could be done but was engineering hell to actually work?'
Yeah.. but Tesla's ideas that people thought were crazy for the most part were. The things that he developed that really became part of the modern world were innovative, but people working in the industry at the time were not all that surprised by them. On the other hand he made lots of fantastic claims that have become part of his modern mythology, but those ideas didn't actually work. He was a dreamer with great intuition, but he lacked a more complete knowledge needed to actually know which of his dreams could work and which did not align with the larger body of physics.
He also appears to be an emdrive proponent, or at least a variation of it.
Unfortunately, it is starting to look like we might be done with breakthroughs. We got a big wave of them when physics went from being mostly wrong to mostly right, but over the last half century we have generally been finding that our knowledge, while still incomplete, is so interlocking and well supported that there is less and less room for something really groundbreaking.
See, this is why publishing even highly speculative papers can be productive. I see this as a response to the paper that everyone keeps talking about in terms of 'omg aliens!', where really the person just layed out how much the object would have to weigh in order for its movement to be described like a light sail and then speculated about what types of objects, such as artificial ones, could have that kind of weight to volume ratio.
So now here we have another person taking that idea and proposing how something with those same characteristics could have a naturally occuring origin. This is what I love about watching scientists figure something out. You start with a mystery, someone proposes a mathematically consistent but kinda out there solution, others pick it apart, work with some bits while dropping others and propose a new solution. This is one of the major reasons publication is so important, so others can build off of ideas, even half baked ones.
Well, 'spaceship' covers a lot of territory, and would be pretty much anything of artificial construction that was not debrits. Such an object could be from 'anywhere', it doesn't have to be somewhere close or only have a short travel time.. a dead probe like Voyager circling around the galaxy a few times would count as a 'space ship', at least in terms of the paper the person was alluding to.
That logic applies to all those other dangerous jobs too. For every person killed in logging, quite a few more are severely injured or permanently maimed.
Yeah.. cops keeping their 'job as safe as possible' makes them a danger to everyone around them, and ironically a danger to themselves since they have been trained to escalate as a first solution, which increases the chances of things becoming violent. Their 'safety practices' are why the job is so dangerous.
Because they keep seeing 'cops are in danger' on TV, and their union bosses tell them 'you are always in danger', then they hire expensive motivational speakers and trainers (who just happen to be ex-cops) to come in and scare all the cops into believing they are constantly in danger and thus they must use excessive force 'just in case'.
Yeah, but just like the big copyright holders, they will probably have special protections put in place for some channels while not others. DMCA takedowns already do not generally work on 'official' channels, there will probably be something similar for this comment policy.
That.. is a really good question. Something as simple as auto-hide comments and have a switch to make them visible or not. But then again, that would be a good solution from the product's perspective, not the customer.
I think a big problem with these questions is not if they are truly hard or difficult, but the meta thinking involved. Many of them really come down to 'which trick did the person who designed this test have in mind?' or even worse 'which niche within the language is the person who wrote this obsessed about?'.
Nah, we here at slashdot love or hate patents based off the story. This is big bad government that should be replaced by private companies vs plucky innovator, so naturally slashdot is generally on the side of the person they can relate to.
Tell that to the patients who have to endure actual government-run healthcare in the US: Vets (the VA) and Native Americans (the BIA). Neither of these institutions have treated their patients with the care that they should have.
Yeah, but those are voting blocks with no power, so they don't really get much priority.
The problem isn't congressmen, the problem is voters. They don't come up with this stuff randomly, they come up with these laws because the primary voters they need to stay in office love them. This type of law exists purely to add to 'things I tried to accomplish FOR YOU' campaign flyers.
True, they struggle to, as you say, deliver results. But what they do produce, you can explain and validate. ML gives you answers, quickly, which is great as long as the answers don't actually matter. That has made it great for recommendations and toys, but is really alarming when applied to more serious stuff.
Moving further away from consumer products and services, there is the worrying limitation of ML that, well, you don't really learn anything from it. Seeing it used in research always worries me since it means the researcher doesn't understand the relationships and is hoping a black box will give them an answer, an answer they can't understand since it can't explain, and no one else can validate. Again, great when you want an answer but don't care what it is, if it is right, or leaning anything from it.
Something I never liked about machine learning, and 'new fangled AI' in general is how opaque it is, you get fast interesting results but you can not explain how you got them or defend them directly. But GOFAI techniques are out of style right now, and it is getting worse as GOFAI systems are so much slower and resource intensive not to mention require so much more domain knowledge to set up and just can not compete with the sexy instant gratification that machine learning can give you.. or give your customers/sponsors.
So.. while not an academic, this is pretty close to my field of research. Looking at their model, I am not surprised they sold this product but deeply disappointed. This is the type of model that is REALLY easy to sell to people, both law enforcement and the military (our customer) are enamored with them for their near magic ability to 'predict' things. Only they don't, they tend to fail in unpredictable ways. They are not bad in multi-model systems where you take a dozen or so different systems built by different teams, run them in parallel, then have subject matter experts ponder the conflicting results. But actual police out of a single model? Madness... or hubris.. or stupidity... or simply being enamored with a slick sales pitch from 'one of your own' offering to solve problems in the way you want them solved.
Oddly enough, we actually DID do a LEO model years back, which was actually pretty effective, but it encouraged things like community outreach and police/citizen interaction which worked really well for officers on the ground but pissed off lawmakers and 'police unions', so it was largely dropped.
Which gets back to this story and one of the fundamental flaws in such attempts. The decision makers are not interested in solutions that make things better for high crime areas in the first place, the people in those areas are not part of their power block. They want solutions that 'sound right' to people who live elsewhere and confirm what they already believe. Which is exactly what models like this are good at producing. They are kinda like torture... useless for prediction or information gathering, but an excellent political tool for confirming the story your career depends on being 'true'.
Slashdot Mirror
I would put it at 'crap shoot'. Journals and patents have the same basic problem of 'depends on who you get', with wildly inconsistent standards, and in both cases they become trivial to handle when you hire on people who's domain is dealing with patents/publications. Same with research grants.
The military always has lots of research going on that is not public, but while it is often technically impressive to people in the relevant fields, it is almost always pretty mundane sounding compared to public imagination.
I would not be surprised if they had some new launch vehicles and better communications/observation equipment than the civilian sector, I would not even be surprised if some department had something like a functional linear aerospike based launch system or some other incremental improvement like that.
But in general, people working in research tend to have a good idea of what the current possibility space is, and secret military projects that come out later are rarely more than 'oh wow,they got that thing we all knew could be done but was engineering hell to actually work?'
While drama is fun, comments like this are one of the big reasons I actually look through threads on slashdot.
Even when patents violate the known laws of physics, patent examiners generally do not have the background to spot these outside free energy devices.
Yeah.. but Tesla's ideas that people thought were crazy for the most part were. The things that he developed that really became part of the modern world were innovative, but people working in the industry at the time were not all that surprised by them. On the other hand he made lots of fantastic claims that have become part of his modern mythology, but those ideas didn't actually work. He was a dreamer with great intuition, but he lacked a more complete knowledge needed to actually know which of his dreams could work and which did not align with the larger body of physics.
He also appears to be an emdrive proponent, or at least a variation of it.
Unfortunately, it is starting to look like we might be done with breakthroughs. We got a big wave of them when physics went from being mostly wrong to mostly right, but over the last half century we have generally been finding that our knowledge, while still incomplete, is so interlocking and well supported that there is less and less room for something really groundbreaking.
See, this is why publishing even highly speculative papers can be productive. I see this as a response to the paper that everyone keeps talking about in terms of 'omg aliens!', where really the person just layed out how much the object would have to weigh in order for its movement to be described like a light sail and then speculated about what types of objects, such as artificial ones, could have that kind of weight to volume ratio.
So now here we have another person taking that idea and proposing how something with those same characteristics could have a naturally occuring origin. This is what I love about watching scientists figure something out. You start with a mystery, someone proposes a mathematically consistent but kinda out there solution, others pick it apart, work with some bits while dropping others and propose a new solution. This is one of the major reasons publication is so important, so others can build off of ideas, even half baked ones.
Well, 'spaceship' covers a lot of territory, and would be pretty much anything of artificial construction that was not debrits. Such an object could be from 'anywhere', it doesn't have to be somewhere close or only have a short travel time.. a dead probe like Voyager circling around the galaxy a few times would count as a 'space ship', at least in terms of the paper the person was alluding to.
That logic applies to all those other dangerous jobs too. For every person killed in logging, quite a few more are severely injured or permanently maimed.
Yeah.. cops keeping their 'job as safe as possible' makes them a danger to everyone around them, and ironically a danger to themselves since they have been trained to escalate as a first solution, which increases the chances of things becoming violent. Their 'safety practices' are why the job is so dangerous.
Because they keep seeing 'cops are in danger' on TV, and their union bosses tell them 'you are always in danger', then they hire expensive motivational speakers and trainers (who just happen to be ex-cops) to come in and scare all the cops into believing they are constantly in danger and thus they must use excessive force 'just in case'.
Police anxiety is surprisingly profitable.
Yeah.. I suspect some horrible mess of nested if-then-else clauses with some fall throughs or cases with errors in them.
Yeah, but just like the big copyright holders, they will probably have special protections put in place for some channels while not others. DMCA takedowns already do not generally work on 'official' channels, there will probably be something similar for this comment policy.
That.. is a really good question. Something as simple as auto-hide comments and have a switch to make them visible or not. But then again, that would be a good solution from the product's perspective, not the customer.
I think a big problem with these questions is not if they are truly hard or difficult, but the meta thinking involved. Many of them really come down to 'which trick did the person who designed this test have in mind?' or even worse 'which niche within the language is the person who wrote this obsessed about?'.
Ah yes, the 'guess which compiler we are using and how it differs in minor ways from the one you know in 20 minutes!' style test.
I am picturing the current crop of 'the past was stupid, never learn from it!' programmers diving into LISP, Smalltalk, or Prolog.
But as always, designing new languages is more fun than learning old ones, and learning from past mistakes makes you uncool.
Nah, we here at slashdot love or hate patents based off the story. This is big bad government that should be replaced by private companies vs plucky innovator, so naturally slashdot is generally on the side of the person they can relate to.
Tell that to the patients who have to endure actual government-run healthcare in the US: Vets (the VA) and Native Americans (the BIA). Neither of these institutions have treated their patients with the care that they should have.
Yeah, but those are voting blocks with no power, so they don't really get much priority.
The problem isn't congressmen, the problem is voters. They don't come up with this stuff randomly, they come up with these laws because the primary voters they need to stay in office love them. This type of law exists purely to add to 'things I tried to accomplish FOR YOU' campaign flyers.
True, they struggle to, as you say, deliver results. But what they do produce, you can explain and validate. ML gives you answers, quickly, which is great as long as the answers don't actually matter. That has made it great for recommendations and toys, but is really alarming when applied to more serious stuff.
Moving further away from consumer products and services, there is the worrying limitation of ML that, well, you don't really learn anything from it. Seeing it used in research always worries me since it means the researcher doesn't understand the relationships and is hoping a black box will give them an answer, an answer they can't understand since it can't explain, and no one else can validate. Again, great when you want an answer but don't care what it is, if it is right, or leaning anything from it.
Something I never liked about machine learning, and 'new fangled AI' in general is how opaque it is, you get fast interesting results but you can not explain how you got them or defend them directly. But GOFAI techniques are out of style right now, and it is getting worse as GOFAI systems are so much slower and resource intensive not to mention require so much more domain knowledge to set up and just can not compete with the sexy instant gratification that machine learning can give you.. or give your customers/sponsors.
RMS has always been about specific freedom for specific people (like him), but not others.
Yeah, those pesky domain experts and their actual knowledge of systems with proper tools of validation. So stupid.
So.. while not an academic, this is pretty close to my field of research. Looking at their model, I am not surprised they sold this product but deeply disappointed. This is the type of model that is REALLY easy to sell to people, both law enforcement and the military (our customer) are enamored with them for their near magic ability to 'predict' things. Only they don't, they tend to fail in unpredictable ways. They are not bad in multi-model systems where you take a dozen or so different systems built by different teams, run them in parallel, then have subject matter experts ponder the conflicting results. But actual police out of a single model? Madness... or hubris.. or stupidity... or simply being enamored with a slick sales pitch from 'one of your own' offering to solve problems in the way you want them solved.
Oddly enough, we actually DID do a LEO model years back, which was actually pretty effective, but it encouraged things like community outreach and police/citizen interaction which worked really well for officers on the ground but pissed off lawmakers and 'police unions', so it was largely dropped.
Which gets back to this story and one of the fundamental flaws in such attempts. The decision makers are not interested in solutions that make things better for high crime areas in the first place, the people in those areas are not part of their power block. They want solutions that 'sound right' to people who live elsewhere and confirm what they already believe. Which is exactly what models like this are good at producing. They are kinda like torture... useless for prediction or information gathering, but an excellent political tool for confirming the story your career depends on being 'true'.