For the first time we compared a CNN’s diagnostic performance with a large international group of 58 dermatologists, including 30 experts. Most dermatologists were outperformed by the CNN. Irrespective of any physicians’ experience, they may benefit from assistance by a CNN’s image classification.
H/t to the AI discussion in SGU#673 (or this link) Every profession will have a tailored AI system related to it in the near future, and this development is a good thing. It’s a lot like Garry Kasparov and the game of chess. These days you don’t become a master at chess without collaboration with an AI system. There is currently no way you can experience the hundreds of thousands of games or examples of whatever your profession does, in the short lifespan of the average human.
We, as consumers of healthcare, should demand that all diagnosis be run through these kinds of systems once they are available. That would just be prudent behavior. Will tailored AI take over the world? Would that necessarily be a bad thing? Don’t make a megalomaniacal AI. That would probably be the place to start; or if we do, we should be sure to not follow it’s advice and to ostracize anyone who does.
That is what Twitter is, a headline writing contest. Anyone who believes it is more than that is deluded.
Twitter is 98% bullshit. That is why the Orange Hate-Monkey thrives there. He’s also 98% bullshit. Second, I’m one of the people who uses another platform (Tumblr) to auto-post to Twitter because Twitter is 98% bullshit and no one with a functioning brain would spend any amount of time on Twitter.
Having said that, there *are* bot swarms on Twitter which make up a huge portion of the traffic on Twitter, and I wish the media types who live on Twitter understood just how meaningless their social platform of choice is. Everyone else is on Facebook. Why aren’t they?
The internet is not just a cheap, fast, and easy way to spread information. It is also a force multiplier. Small information campaigns can end up having a massive effect, for two important reasons. One is that the inherent structure of the web allows for and encourages the spread of information. Some kinds of information spread faster and wider than others. So we need to ask ourselves – what features of information will make it spread more through social media? It’s not accuracy, or thoroughness, or fairness. Bite-sized nuggets of drama or humor seem to do the best. If your information is unencumbered by reality, that is an advantage.
I have known for a long time that you have to believe a thing before you can convince others. I see no reason to limit that observation of human nature by saying the things you manipulate with must be true. They just have to sound true, and most of that is presentation.
It’s good to get confirmation of the hypothesis through scientific study, even though I’m not surprised by the results. Most people will only watch/read far enough to get confirmation of the thing they are told to believe, want to believe. It takes effort to discover the whole picture. If there is no reward in filling out the picture, most people will not bother.
I would be interested in discovering which portion of the test group refused to do the work after watching all the videos. Were they less inclined to demonize or praise? That information would be more revealing of human nature, in my opinion.
Skeptics Guide to the Universe #607 – Science or Fiction. Steve used the word only to describe capturing data from the LED on the computer. Using malware in addition to the LED is not only using the LED. If you add the variable of freighted malware to the possibilities, there are many ways that information from your computer could be gathered, even if it was never connected to a network.
The only way to be certain that your code is clean is to write every bit of the code yourself on an isolated machine, and even then malware can be encoded onto the ROM chips built into the board.
Basically, if you want certainty that your systems are secure, you are screwed. But malware in addition to the LED is still not only using the LED.
Written reference to the superstitious fear of the number thirteen dates to the late 1800s. Its origin is conjectural (a matter of guesswork). The term triskaidekaphobia first appeared in the early 1900s. It was derived from treiskaideka, the Greek word for thirteen + phobia, fear of = a fear of thirteen.
Thirteen is supposedly a bad number because the twelve disciples plus Jesus equals thirteen, the first reference that she offers for the fear of that day and/or number. I hadn’t heard the cycles (moon, menstrual) argument before. I have never (and I do mean never) heard the triskaidekaphiliac women’s day argument before.
The thirteenth is my lucky day. I was born on the thirteenth. I got married on the thirteenth because the wife insists I remember things that fall on the thirteenth day of the month. She also scheduled the births of our children (C-sections are like that) for the thirteenth of the month. It isn’t her fault the children didn’t actually emerge on those days (birth is like that) So when Friday the thirteenth rolls around I enjoy the double-whammy of good luck; my favorite day of the week and my favorite day of the month combined into one great day to celebrate. I am the biggest promoter of triskaidekaphilia that I know of aside from this guy.
What I’m trying to say is I of all people should have heard the women’s day argument before, and I haven’t. So I’m going to say Friday the thirteenth being a women’s day is the fiction. Hope that clears it all up for you.
The Skeptic’s Guide to the Universe #577 is out. While listening, I was agog at the notion that scientists could, with a straight face, attempt to determine if cool was a real thing. This is the kind of thing a nerdy scientist thought that they might be able to prove.
It’s a game of superficially rebellious status-chasing, centered on consumerism.
What then followed was Cara Santa Maria trying to convince the rest of the panel that there was a difference between being edgy and posing. All of that is coded language. Cool is coded language. Cool used in any other way than to describe the temperature of something is creating a meaning for the word other than what the word really means, the relative lack of warmth of an object or area. Cool has no meaning aside from temperature unless you want to try and find the shallowest surface of existence, to give deep meaning to what marketing types want to try to sell you next.
Edgy is always posing. People who get tattoos and piercings because they want to be seen as cool or on the edge really don’t understand what damage that kind of real life living entails. When your normal day involves scrounging food out of a dumpster, scoring a high so that you can get through another suck-ass day, those scars paint themselves on you without you having to go looking for the edge of normalcy.
When you live that kind of life on a day to day basis, you aspire to be normal. You aspire to have what some external observer might call a normal day. Meals prepared from food you bought in a store, eaten at a table with real chairs, with all the family present. Getting through a day without breaking things in anger. What a relief normalcy can be, when normalcy is something that you just vaguely remember seeing on TV once, a long time ago.
You pulled a fast one this week, Steve (Skeptic’s Guide to the Universe #543) I’ve written posts for my blog on the subject of Santa Claus, and I even reference the same source material that Snopes.com references for their article. While it may be strictly true that Coca-Cola did not ‘create’ the modern image of Santa Claus with a specific campaign, they have shaped the image of Santa Claus with their promotions over the years. So much so that the image most people have for Saint Nick is the Coca-Cola image specifically.
I dispute both Snopes’ ruling of false and your using this as the fiction for this weeks SGU podcast on that basis alone. The article on Snopes’ site might be more persuasive if the example images for their arguments actually loaded. But I doubt it.
I agree there is ambiguity in saying that Coca-Cola created the modern image. They contributed to an iterative process, perhaps substantially.
That is exactly why I included the additional line that prior to Coke the image of Santa was tall and thin. That part made it unequivocally fiction. The elements of the modern Santa predate Coke. Coke just put together one artist’s conception of Santa.
…as I said previously, the notation that he was ‘tall and thin’ previous to the Coca-Cola campaign being the fiction would carry more weight if the Snopes article actually loaded the photos of what Santa looked like before the campaign. As it is I have to go find images myself, and having dived into this subject a few times now I have to say that research on this subject is made more difficult by the number of people who seem to think they know something about the subject writing contradictory articles about it.
Like the subject of Christmas (or Yule) itself, separating fact from fiction is a laborious process.
(feedback exchange with the SGU)
Editor’s note. In this exchange of messages with Dr. Steven Novella I failed to mention that both my grandmother and her sister had wood carvings of Santa Claus that they displayed during the Christmas season. In both cases the carvings looked very old and they were both of a tall and thin bearded man. These two anecdotal experiences do not equate to a general pattern of perception, but it does lend weight to the belief that one might have about Santa Claus as being a tall, thin man if you had seen nothing but these kinds of icons representing Santa in your family.
I still dispute the finding. I’ve never gotten the page at Snopes.com to load properly until today. The variance in the red costume, wide belt, whiskers and ruddy complexion are negligible. So Coca-Cola did not create the image. Fine. Some of the images are not of a grossly fat man, though. This confuses the question as presented by Steve in that Science or Fiction. My verdict stands.
One of the most widely accepted conspiracy theories in the US remains the assassination of President John F. Kennedy. Just last week I heard someone suggest that Oswald didn’t act alone. Statistics show that more than half of US residents agree with this statement, and are convinced to this day that Oswald was a patsy, silenced by Jack Ruby a few days after the assassination.
For many, many years I was one of those people. I read several books on the subject, I watched every documentary, I even went to Dealey Plaza once simply to stand next to the spot where Kennedy was shot. In many ways the assassination of JFK was the lynchpin for all of my conspiratorial thinking; it was the first conspiracy theory I had ever heard, it was the most solidly defensible of any of the many popular conspiracies that cropped up later (so much so that even the US government has agreed there was a conspiracy, contradicting the findings of its own commission that investigated the assassination) and once I was led to question that theory, my belief in all those other theories also crumbled.
Why shouldn’t they, when they didn’t even have a magic bullet to hide behind?
The trip to reality was long and arduous for me. It started about the time I started writing this blog, and continues to this day. Every single thing I read these days sends me off looking for corroborating sources and counter-arguments, just so that I can be sure I’m dealing with real facts and not some fever dream of the magical thinking majority.
I wish I had access to Case Closed when I was a young man looking for facts on the JFK assassination. The depth of investigative research that Gerald Posner has gone to is unequaled amongst the many different authors on the subject. Here is an interview with Posner from 2013, discussing the mountains of evidence linking Oswald to the killing, and detailing the kind of man Oswald was.
If Case Closed had been available to me when I first started looking into this subject, I never would have started down that rabbit hole of conspiratorial thinking in the first place. Would have simply come to the conclusion ah, Oswald shot Kennedy and left it at that. But I didn’t have access to that book back in the 70’s when I was into the subject. I don’t even remember the titles of the books I did read; but I do remember The Men Who Killed Kennedy documentary being something I watched and rewatched many times, as well as the Oliver Stone film JFK which I remember receiving quite credibly.
Except for one thing. The repeated mantra back and to the right which Stone puts in Garrison’s mouth in the film. I actually went back and reviewed the Zapruder film because of this, and discovered that the motion he insists is there really isn’t there at all. The film clearly shows the headshot coming from the back and above, just as Posner says in the video.
But I didn’t have Posner. Never ran across his book until recently, while listening to back episodes of the SGU (like so many good skeptical habits I have picked up) what I had was my own inability to ignore evidence when it is presented to me. What I stumbled across was this re-enactment (one of several) proving that the magic bullet was nothing of the kind. That the trajectory of the bullet is mappable and repeatable given an accurate reproduction of the events of that day.
The second source of video was a very detailed recreation of the exact poses of the victims taken from Zapruder film footage, that were mocked up by Anatomical Surrogates Technologies for the documentaryJFK: Beyond The Magic Bullet . (full video available in three parts here)While the shot does appear to strike too low, the trajectory is almost identical to the bullet on that fateful day.
Lastly we have the recreation of the headshot showing that the direction that Oswald fired from was indeed the only direction where the damage seen to the President’s head can be replicated. For those who simply aren’t convinced by the replication of the magic bullet’s trajectory.
Conspiracy theorists will of course come up with reasons why this proves nothing. Personally I see no reason to continue pretending that Oswald did not kill Kennedy. If you feel the forensic tests are simply not enough evidence, then I encourage you to pick up a copy of Case Closed. If none of this suffices, then I suggest you look to your own mental barricades. If your beliefs cannot be falsified, it says as much about your failings as a critical thinker as it does the indefensibility of your opinions.
As Dr Novella goes into on his blog entry, conspiracy theorists attempt to discredit evidence that would seem to destroy their preferred fantasies by picking apart the details of the evidence, looking for the slightest anomaly that they can then use to discredit it.
Having watched The Men Who Killed Kennedy I remember the attempts to discredit this photo and the autopsy photos quite vividly. I remember wondering at the time why anyone would go to such lengths to hide evidence, marveling at the scale of the conspiracy required to perpetrate such a massive hoax.
It is with a wry chuckle that I remember my own gullibility on the subject. The understanding of the scale of the conspiracy should have been my first clue as to the implausibility of the conspiracy itself. That understanding would take years to mature, though.
The computer simulation embarked upon to validate this photo is as much of an over-the-top effort to show the solidity of the evidence for Oswald being the shooter, as the series of videos I linked above was. In the study linked here, you can see the many points of data used to determine if Oswald is actually standing in a stable position, and that the shadows in the photo match the shadowing that would have been present at that time of day and season of the year.
This is the kind of thorough analysis that is required to refute the claims of conspiracy fantasists who continue to insist that it simply wasn’t possible for such a insignificant little man to have killed the most powerful man in the world single-handedly. At least the computer modeling techniques showcased here can be used for many other instances of questionable photographic evidence, so that their validity can also be certified.
Originally posted here, radically enlarged and embroidered here, I’ve copied this to the day it should have been published on, and will be published on in the future if the wild hair suits me. Conspiracy fantasists are getting on my nerves these days, and I don’t feel like cutting them any slack.
97% of the state of California is currently experiencing a drought, according to the website Californiadrought.org. 46% of the state is suffering in an “exceptional drought”, the most severe status. California’s central regions, it’s farmland, is taking it on the chin, but overall, there is estimated to be about 36 million Californians effected by this drought. Any way you look at it, the water situation in California is dismal.
I had relatives who dowsed. They believed in their abilities to find water and would deny that what they engaged in was magical thinking. On the other hand, I have no way to explain how they found water; and they did find water, in some cases more reliably than engineering firms of the time. Sadly none of them are still breathing, so I can’t test them or suggest they submit themselves to testing. I simply resent the blanket dismissal of ‘magical thinking’ as it relates to the subject of dowsing.
A farmer knows his own land, sometimes in ways that he has a hard time explaining. Finding water on land that he is familiar with might actually tap into understanding about the lay of the land, the various structures that would lend themselves to collecting and channeling water. The act of dowsing then simply gives form to the assembling of knowledge already at hand. A method for explaining why they know where water is most likely to be.
All of this is a completely different subject than the notion that you can pay someone else to dowse your property and get a return on your money.