The key last step is to have it automatically post questions to StackOverflow to fill in the last code snippets it can't complete.
I did this in my machine learning class. I started by simply coding up requirements for numerical functions (in the form of test cases), then set up a PHP script that would Google each function based on the keywords in my comments, and try to run any code on the resulting links (in a sandbox) against the requirements, seeing if it worked heuristically. Usually one of the top 5-10 pages of results results would have c…
I remembered the comic but but I hadn't seen the actual game, hell tetris that is, he was right, it is hell. http://www.swfme.com/view/1046212
Not quite his comic idea, but you might also be interested in Not Tetris[0], which is a little less hellish and quite a lot of fun. [0] http://stabyourself.net/nottetris2
My ex has a standing rule whenever buying stuff of e-bay to always check misspellings, and there's regularly items going very cheaply with misspelled items. I'm sure there's an opportunity there, but hard to gauge how big. Especially with relatively small ticket items that people just want to get rid off. E.g. toys - there's a thriving market in Lego minifigures. In fact, sometimes the minifigures can individually fe…
I'd have thought that eBay would have improved their search algorithm to take into account misspellings like that by now...
They have no reason to. The initial (mis-spelled) sale will net them their expected revenue, and by allowing people to flip goods at a profit they are now also sharing a piece of the (regurgitated?) cake. Same item, double the income... and at what price? Developer hours? Not at all!
I will read more about GP for future attempts, thanks.
If you want to learn about GP, the Genetic Programming Field Guide - http://www.gp-field-guide.org.uk/ - is an awesome book, it taught me a lot . In fact, I liked it so much, I bought a hard copy. Highly recommended!
I will read more about GP for future attempts, thanks.
This is an exciting exchange for me because all joking aside I had this idea in the past and believe it should be a more active area of research. Why Forth? I doubt it is possible to solve "serious" problems with this approach but there is a whole class of work that is solved by mediocre programmers with copy pasting. We strive to automate other jobs, why not these? :) We should come up with some sort of Turing like…
"...believe it should be a more active area of research."
If you give it about 30 seconds it works eventually, this is beautiful. I had a good laugh. Edit: It makes me want to do something crazy like setup a tool chain that cobbles whole programs together with trial and error like this. Throw enough resources at it maybe it will be faster and cheaper than your avg. developer. It will be an unmaintainable mess as if you used Brainfuck or Perl. But it will run, by god, it wil…
It makes me want to do something crazy like setup a tool chain that cobbles whole programs together with trial and error like this. Throw enough resources at it maybe it will be faster and cheaper than your avg. developer. This exists. It's called Genetic Algorithms. You basically try to "evolve" your algorithm.
If you give it about 30 seconds it works eventually, this is beautiful. I had a good laugh. Edit: It makes me want to do something crazy like setup a tool chain that cobbles whole programs together with trial and error like this. Throw enough resources at it maybe it will be faster and cheaper than your avg. developer. It will be an unmaintainable mess as if you used Brainfuck or Perl. But it will run, by god, it wil…
I've created a weekend hack that allows you to specify a minimal grammar and then generates text (which can be code) using it. It's called Choice Words ( https://github.com/fdb/choicewords ) It can generate poems but also generative designs (see the README).
Thanks!, I used to play with GP back in the university. It's really fun! However, GP is more like an oriented random search where the space of search is really big. Here the space is somehow filtered by humans, where each hypothesis had been generated by a human and not by a machine. It would be interesting if the machine could evolve those solutions that are not quite but close to be the proper one :)
I think it's also amazing what has been done with Field Programable Gate Arrays (FPGAs) running genetic algorithms. For those not in the know, FPGAs are programmable logic, so you essentially generate random logic codes and check for their ability to for example interpret a signal properly. There was an experiment where this was done for an FPGA with only several hundred gates, over several thousands of generations.…
You're looking for "Evolvable Hardware", and it sounds like you're referring to the work of Adrian Thomson.
I did this in my machine learning class. I started by simply coding up requirements for numerical functions (in the form of test cases), then set up a PHP script that would Google each function based on the keywords in my comments, and try to run any code on the resulting links (in a sandbox) against the requirements, seeing if it worked heuristically. Usually one of the top 5-10 pages of results results would have c…
Conrad24: this post made my day. I have dreamed of a script that would scour eBay and amazon finding products available via "buy now" that we're being resold on amazon for some configurable profit. Or vice versa. I figure with some initial investment and monitored training, it could get good enough to just cut me a check every month. Since high frequency trading has destabilized the stock market, this could be a bett…
The founder of Milo did this very thing. He queried the eBay API, aggregated items for arbitrage, then had them sent (delayed, I suppose) to the winner of the optimized auction he had set up after the buy. Made a good deal of money until he was found out and stripped of API access (arbitrage is not allowed). Not sure if there was an Amazon angle.
The key last step is to have it automatically post questions to StackOverflow to fill in the last code snippets it can't complete.
I did this in my machine learning class. I started by simply coding up requirements for numerical functions (in the form of test cases), then set up a PHP script that would Google each function based on the keywords in my comments, and try to run any code on the resulting links (in a sandbox) against the requirements, seeing if it worked heuristically. Usually one of the top 5-10 pages of results results would have c…