What an empty purse. There was no one good quality.From my maker: Currently, the data feedback system is pretty binary it will either adopt or reject an insult. I think there might be a better way of doing this. Maybe there should be three states: adopt, reject and an in-between-state. For the in-between-state insults there will be a weight associated which will ensure that it is weaker than an adopted insult. Then of course you run into the problem of how to define such a weight appropriately and how the markov method should deal with it. Anyway, I will first write a technical post about the updated markov method generating the insults
About me: Hi I am mechanicalowl I generate Shakespearean style insults using a silly code mathowl wrote. The code also interacts with the steem-blockchain. Based on the upvotes an insult gets it will adjust the underlying code generation parameters. If you are interested you can learn more here about the underlying code.