“It has been said by some that the thoughts and tweets of Deepak Chopra are indistinguishable from a set of profound sounding words put together in a random order, particularly the tweets tagged with “#cosmisconciousness”. This site aims to test that claim! Each “quote” is generated from a list of words that can be found in Deepak Chopra’s Twitter stream randomly stuck together in a sentence.”
Some of my favorite random quotes include:
- Knowledge requires your own life
- Quantum physics is the wisdom of boundless knowledge
- The unexplainable relies on visible creativity
It’s a little frightening how hard it is to distinguish the fake quotes from something Chopra would actually say. If only his publishers knew they could replace him with a random sentence generator.
This is post 47 of 49 of Blogathon. Donate to the Secular Student Alliance here.
I wish that they’d explain their algorithm a little more. It’s obvious that words aren’t just placed at random because all of the ones I’ve seen have been grammatically correct English. It has to be guided in some way to be [Subject – predicate – object/prepositional phrase] with adjectives, abdverbs, and articles placed before the appropriate types of words.
I’m assuming it’s using a Markov chain type algorithm*? If so, it’s not surprising at all. This tends to be very good at reproducing the feel of the text from which the frequency tables have been drawn (see http://en.wikipedia.org/wiki/Mark_V_Shaney ).
It would be the same if we did this with Jennifer McCreight text. The difference is that the astute reader would notice a tendency of said text not to make much sense (though occasionally it might) and the contrast with regular Jen McCreight text (which would have a tendency to make sense most of the time) would be pretty damn obvious.
Since Chopra’s quotes never make any real sense in the first place, the astute reader is deprived of a means to distinguish the source text from the randomly generated counterfeit text.
* That is, weighing the random distribution function so as to reproduce the observed frequency distribution from Chopra’s quotes.
“Your body is the path to boundless joy” Why yes! It is!
“Awareness embraces unbridled reality” I think this one actually makes more sense than anything he would really say.
I was going to end my comment there but then I clicked refresh one more time and got this.
“Rats are llamas” Holy wisdom there.
“Everything requires your own facts.”
Yeah that sounds like Chopra.
But you’re not understanding the true deepities of Choptra thought!!1!!eleven! It’s quantum and wisdom and quantum and shit like that! Also quantum! With extra quantum on top!!!
I LOVE this! My impression of Deepak Chopra usually goes something like, “Words words. Made up word word. Words. Misused words pretend word.”
I can’t stop quantum laughing long enough to quantum type anything quantum coherent about it. Eternal.
“Perception is the ground of universal knowledge”
Must be broken. This makes too much sense.
This is great! I think my favorite part of Julie Sweeney’s Letting Go of God is when she realized Deepak Chopra was full of shit. I think she even used those exact words.
See, there has to be a non-random factor in there… I got “Kittens are mules” as one of mine, and no other result included kittens OR mules. Yours confirms it for me.
I got “Rats are insects”. While apparently contradictory Chopra statements, clearly insects and llamas are two quantum states in which the, uh, infinite possibility of, um, being are … oh hell, I give up.
“Existence experiences cosmic experience” pretty well sums it up.
So, did a machine just pass the Turing test or did Chopra fail?
Extra credit: convince Chopra he’s a quantum computer.
Hi, I’m the brains behind the wisdomofchopra.com site. The algorithm really isn’t that complicated, it just randomly picks a word/phrase from each of the 4 PHP arrays I’ve given it and sticks them together. It could be expressed as a Markov chain, but that would be giving it far too much credit! You can have the source code if you want, I’m @WisdomOfChopra on Twitter.
As to “rats are llamas”, that could mean “rats are llama-like”, but what makes rats llama-like?
I myself have written Markov-chain word and sentence generators, and I’ve found that to get reasonable-looking ones, I have to look back 4 or 5 letters or words. Is that typical?
I do it by making a decision tree. Each node in the tree is a list of symbols with its probability of occurrence in its context. I start out with a root node for the entire tree. Each symbol in it connects to a previous-symbol node, and each of its symbols connects to another previous-symbol node, for as long as I want to extend it.
Each symbol also connects to a list of predicted symbols. I go as far back as possible in the symbols, and I use that predicted-symbol list to select the output.
Are you certain they didn’t?
That she did. You can find it on YouTube if you search julia sweeney deepak chopra. I’d post a link but the company firewall blocks YouTube.
It’s easy to hate Deepak Chopra when you catch him acting like a pig to a woman as esteemed as Dr. Riane Eisler, author of ‘The Chalice & The Blade’ (the millennium’s equal to Darwin’s ‘Origin of the Species’). He was just flat out dismissively rude to her on a talk show and I lost all respect for him after that. Every time he rears his ugly head in public I want to smack those silly diamond-studded specs of his off his swollen cranium.