Note that I am only considering the analysis of 2D images, and not "direct" 3D info from stereoscopy (multi-angled imaging), laser ranging, etc. as often found in self-driving cars.
Humans generally do 3D mental modelling and find the best model to explain the observation. If we see a cheetah-like thing, we try to figure out how it's oriented: where are the legs, the head, the tail, etc. We expect cheetah's to have to have those and thus look for them.
If presented a NON-animal with a cheetah pattern, we'll have trouble finding expected cheetah parts, which makes us re-evaluate the animal assumption. It has a cheetah pattern, but the 3D shape of it more resembles a couch than an animal and thus we test the "couch theory" next.
Most of this is subconscious, unless we have trouble making it out (can't find a plausible model) such that we have to consciously ponder alternatives. "Maybe it's a fat dog-bone with a cheetah pattern on it? Maybelle, what do you think?..."
I suspect better AI will have to do similar 3D modelling to test the plausibility of the model against the actual observation. Neural networks (NN) as currently implemented are not sufficiently capable of that. A good modelling system would be able to produce a 3D model of the subject (original image) along with the lighting direction/type assumptions such that if one renders the model, it produces a close match to the original (target) image.
But even that may have limits. For example, humans can look at a rendering or drawing with somewhat "wrong" (inconsistent) shadows, and still be able to figure it out. A "pure" model comparison would fail because no "logical" lighting would fit. It would have to accept possibilities of local or distorted lighting. Similar goes for complex shadows, say from spotty tree leaves.
Perhaps genetic algorithms guided by NN guesses will be needed to construct such models for evaluations. It would need 3D sub-models of everyday objects to compare against.
A good ALT attribute shouldn't prevent you from using text on images.
For whatever reason, the tool dinged points for embedded text even if there was an ALT attribute. One can argue using the ALT attribute might be an extra step for a sight-impaired person, and that's why points were dinged. I have other gripes about the automated tool.
But either way, the PHB was viewing the score like an accounting report and we grunts are stuck trying to make the numbers better. To be fair, the PHB has to report concise and objective evidence that accessibility was being improved, and long-winded technical discusses don't work on summarized progress reports up the chain of command. Politics, both public and office, prefer simple messages over rational by long messages.
The article's note about how image identifiers can be "tricked" reminds me of an actual incident.
Our org subscribes to an automated ADA (accessibility) scoring tool. The tool recommends one not embed text into images, rather to use direct text (in HTML). Thus, if it finds text embedded in images, it flags it with a warning and reduces the accessibility score.
Our local PHB looks at the report, and wants a better numerical score. But, the tool was mistaking someone's belt-buckle in an image as text, and marking it on the review report. It appears the tool uses AI to judge if a given image has text in it. I don't directly control the image decisions, so it created a bit of organizational tension.
The belt buckle indeed does kind of resemble text, but explaining why the software is doing this to PHBs can make for some odd conversation. "What do mean the computer guessed wrong? Computers guess?"
I had to find and show the software's disclaimer that said something like, "Each situation should be inspected and judged by a trained (human) accessibility professional, for the report cannot replace the judgment of such professional."
PHB: "So we paid all that money for a stupid computer?"
Me: "Uh, I didn't pick it." (holding my tongue about how they were duped by slick sales-people, as usual...rinse, repeat)
I too had an unpleasant mishap and close calls while a driving newbie. But it's not really comparable because AI cars are trained heavily on private (closed) test courses first, for far longer than a typical middle-class person could afford.
One advantage of AI over people is that you train one copy heavily and thoroughly, then clone it.
Trump broke [partisan payback] cycle...Trump is a hero.
Trump breaks everything. If his chaos fixed something, it was accidental, like Scooby Doo inadvertently crashing into the hologram machine, exposing the ghost as a fake.
his networth tank over the course of the election. There is no way you can say that he did it for personal wealth.
He may not have predicted that up front. He's been reluctant to give up control of his biz's.
You seem to assume too much planning and purpose to his actions. There is no solid record of him carefully planning anything anywhere. His family does most the real work.
find some common ground... like the fact that Hillary and The Donald are BOTH scumbags who don't deserve...
Almost ALL politicians are scumbags. Non-scumbags rarely last in the system; politics is a mean biz. We've been electing scumbags from the beginning. Plus, you need scumbags to understand international scumbags, like angry dictators.
This election was between a sane scumbag and a nutzo xenophobic scumbag.
Hillary didn't bother with any towns smaller than half a mil or so
She was just trying to campaign efficiently, not thumb her nose at small towns. Democrat policy is NOT anti-middle-America (unless you still believe trickle-down works; another debate).
Not that the Orange Overlord [cares about swing states] either, but at least he pretended to
That's the key, he faked it better. He's a sales whiz, she's not. The truth is that automation is a bigger (and growing) threat to the rust belt than bad trade deals. Education is better way to stay ahead of "the robots" than trade gambles, and H had better education policies. (Robots will threaten Chinese factory jobs also soon.)
T gave them at least a lingering hope that reworking the trade deals will bring back old-style jobs. Yes, most voters probably realized it's a long shot, but they were willing to take that gamble. The casino salesman knew how to sell a good gamble. "What've you got to lose?" was one of his favorite phrases.
Bernie is probably pretty close to the humble side...If the dem's handn't shot him down and let him win. I don't think we'd have trump in the big house
Somebody who accepts the "socialist" label is not going to attract centrists, and it's really hard to win without centrists. (I realize there's nuance to such labels, but nuance is often lost in campaign rhetoric. "Socialist" is a 4-letter-word in this country.)
large number of people out there that don't want a bunch of gangsters like the clintons in office
I think what you mean is they took FULL advantage of the grey areas of the law. (So does Trump; he even bragged about such.)
If they survived 50 or so dastardly accusations without a single jail-worthy smoking gun, they are either genius criminals, or the victims of political exaggeration. Occum's Razor says the latter.
Jay Miner seemed like the kind who wanted a real salary, which Jobs couldn't pay. Woz saw the microcomputer biz as more of a hobby. It's why he wouldn't let Jobs slice out the Apple expansion slots to save some dimes.
The opposite may also be the case. We'll never know: God doesn't let us play with his sims. Something tells me Jobs would eventually wind his way into a similar industry and make some waves, but perhaps not as much because Apple gave him early experience to learn from. Jobs had to stumble around a bit before he found his footing: reality is a strong teacher. Without the early Apple experience, he may still be in the stumbling stage.
I took $25/hr after the dot-com crash just to survive because the West Coast was flooded with out-of-work developers. My skills in legacy products saved my arse because web-only newbies didn't have those skills. IT pays well now, but keep in mind bleep happens and the future is unknowable. War, tech bubbles, recessions, offshore-enabling tech, crazy leaders, etc. may slap IT hard. Have a rainy-day fund.
So it would seem deep understanding of the language and the ability to troubleshoot bad code in said language is indeed the crux of the issue.
It could be both. Bad code can be written in ANY language such that COBOL itself is not necessarily the direct problem, but rather mis-use of COBOL by the original-but-gone staff. If the original code was in say Java or Ruby, it too could have been sloppy code, leading to the same problem.
And if there are no domain (subject) experts around, then the organization may have no choice but to hire experts in COBOL but who are not domain experts (or at least don't know that particular org even if they know the general industry).
Slashdot Mirror
Also known as broken Windows economics.
Because they have a near monopoly on biz IT and therefore don't give a flying fock.
Same reason Google sold user privacy to the wolves: they have a near monopoly on searching and therefore don't give a flying fock.
Same reason big telecoms suck in your area...
Not quite sure what that is, but it sounds fun
Note that I am only considering the analysis of 2D images, and not "direct" 3D info from stereoscopy (multi-angled imaging), laser ranging, etc. as often found in self-driving cars.
Explains your hard-drive
Humans generally do 3D mental modelling and find the best model to explain the observation. If we see a cheetah-like thing, we try to figure out how it's oriented: where are the legs, the head, the tail, etc. We expect cheetah's to have to have those and thus look for them.
If presented a NON-animal with a cheetah pattern, we'll have trouble finding expected cheetah parts, which makes us re-evaluate the animal assumption. It has a cheetah pattern, but the 3D shape of it more resembles a couch than an animal and thus we test the "couch theory" next.
Most of this is subconscious, unless we have trouble making it out (can't find a plausible model) such that we have to consciously ponder alternatives. "Maybe it's a fat dog-bone with a cheetah pattern on it? Maybelle, what do you think?..."
I suspect better AI will have to do similar 3D modelling to test the plausibility of the model against the actual observation. Neural networks (NN) as currently implemented are not sufficiently capable of that. A good modelling system would be able to produce a 3D model of the subject (original image) along with the lighting direction/type assumptions such that if one renders the model, it produces a close match to the original (target) image.
But even that may have limits. For example, humans can look at a rendering or drawing with somewhat "wrong" (inconsistent) shadows, and still be able to figure it out. A "pure" model comparison would fail because no "logical" lighting would fit. It would have to accept possibilities of local or distorted lighting. Similar goes for complex shadows, say from spotty tree leaves.
Perhaps genetic algorithms guided by NN guesses will be needed to construct such models for evaluations. It would need 3D sub-models of everyday objects to compare against.
For whatever reason, the tool dinged points for embedded text even if there was an ALT attribute. One can argue using the ALT attribute might be an extra step for a sight-impaired person, and that's why points were dinged. I have other gripes about the automated tool.
But either way, the PHB was viewing the score like an accounting report and we grunts are stuck trying to make the numbers better. To be fair, the PHB has to report concise and objective evidence that accessibility was being improved, and long-winded technical discusses don't work on summarized progress reports up the chain of command. Politics, both public and office, prefer simple messages over rational by long messages.
The article's note about how image identifiers can be "tricked" reminds me of an actual incident.
Our org subscribes to an automated ADA (accessibility) scoring tool. The tool recommends one not embed text into images, rather to use direct text (in HTML). Thus, if it finds text embedded in images, it flags it with a warning and reduces the accessibility score.
Our local PHB looks at the report, and wants a better numerical score. But, the tool was mistaking someone's belt-buckle in an image as text, and marking it on the review report. It appears the tool uses AI to judge if a given image has text in it. I don't directly control the image decisions, so it created a bit of organizational tension.
The belt buckle indeed does kind of resemble text, but explaining why the software is doing this to PHBs can make for some odd conversation. "What do mean the computer guessed wrong? Computers guess?"
I had to find and show the software's disclaimer that said something like, "Each situation should be inspected and judged by a trained (human) accessibility professional, for the report cannot replace the judgment of such professional."
PHB: "So we paid all that money for a stupid computer?"
Me: "Uh, I didn't pick it." (holding my tongue about how they were duped by slick sales-people, as usual...rinse, repeat)
I too had an unpleasant mishap and close calls while a driving newbie. But it's not really comparable because AI cars are trained heavily on private (closed) test courses first, for far longer than a typical middle-class person could afford.
One advantage of AI over people is that you train one copy heavily and thoroughly, then clone it.
Freud: "Female robots have bolt-envy"
I'm not sure "it's no difference". For one, there's a point of diminishing returns.
Trump breaks everything. If his chaos fixed something, it was accidental, like Scooby Doo inadvertently crashing into the hologram machine, exposing the ghost as a fake.
He may not have predicted that up front. He's been reluctant to give up control of his biz's.
You seem to assume too much planning and purpose to his actions. There is no solid record of him carefully planning anything anywhere. His family does most the real work.
Almost ALL politicians are scumbags. Non-scumbags rarely last in the system; politics is a mean biz. We've been electing scumbags from the beginning. Plus, you need scumbags to understand international scumbags, like angry dictators.
This election was between a sane scumbag and a nutzo xenophobic scumbag.
She was just trying to campaign efficiently, not thumb her nose at small towns. Democrat policy is NOT anti-middle-America (unless you still believe trickle-down works; another debate).
That's the key, he faked it better. He's a sales whiz, she's not. The truth is that automation is a bigger (and growing) threat to the rust belt than bad trade deals. Education is better way to stay ahead of "the robots" than trade gambles, and H had better education policies. (Robots will threaten Chinese factory jobs also soon.)
T gave them at least a lingering hope that reworking the trade deals will bring back old-style jobs. Yes, most voters probably realized it's a long shot, but they were willing to take that gamble. The casino salesman knew how to sell a good gamble. "What've you got to lose?" was one of his favorite phrases.
I picked Area 51, but some strange people in suits came to my door, smashed my laptop, and demanded I give them the area.
I don't think so. We might be monogamous on paper, but not in practice. And the "top dogs" still gets more "hot stuff". Even more so under polygamy.
yes, but culture is shaped by our ape-like instincts.
Somebody who accepts the "socialist" label is not going to attract centrists, and it's really hard to win without centrists. (I realize there's nuance to such labels, but nuance is often lost in campaign rhetoric. "Socialist" is a 4-letter-word in this country.)
I think what you mean is they took FULL advantage of the grey areas of the law. (So does Trump; he even bragged about such.)
If they survived 50 or so dastardly accusations without a single jail-worthy smoking gun, they are either genius criminals, or the victims of political exaggeration. Occum's Razor says the latter.
Jay Miner seemed like the kind who wanted a real salary, which Jobs couldn't pay. Woz saw the microcomputer biz as more of a hobby. It's why he wouldn't let Jobs slice out the Apple expansion slots to save some dimes.
The opposite may also be the case. We'll never know: God doesn't let us play with his sims. Something tells me Jobs would eventually wind his way into a similar industry and make some waves, but perhaps not as much because Apple gave him early experience to learn from. Jobs had to stumble around a bit before he found his footing: reality is a strong teacher. Without the early Apple experience, he may still be in the stumbling stage.
Obligotary movie meme.
I took $25/hr after the dot-com crash just to survive because the West Coast was flooded with out-of-work developers. My skills in legacy products saved my arse because web-only newbies didn't have those skills. IT pays well now, but keep in mind bleep happens and the future is unknowable. War, tech bubbles, recessions, offshore-enabling tech, crazy leaders, etc. may slap IT hard. Have a rainy-day fund.
It could be both. Bad code can be written in ANY language such that COBOL itself is not necessarily the direct problem, but rather mis-use of COBOL by the original-but-gone staff. If the original code was in say Java or Ruby, it too could have been sloppy code, leading to the same problem.
And if there are no domain (subject) experts around, then the organization may have no choice but to hire experts in COBOL but who are not domain experts (or at least don't know that particular org even if they know the general industry).
It's a "fake word" such that spelling is mostly moot. Anyhow, here's a better description of the concept.
Without Woz's technical expertise and experience, Apple would probably never exist. It took teamwork.
But Woz is probably not a good example because he's not cut out to be an executive, period. It's a false dichotomy.
Darwin E. Smith or Colman Mockler are better examples of humble yet successful leaders.
Colman M. Mockler Jr., former CEO of Gillette (now part of P&G).
He was very effective, yet very humble.