Comment by rafiss on Harry Potter and the Methods of Rationality discussion thread, February 2015, chapter 108 · 2015-02-23T01:11:28.883Z · LW · GW

There are at least two alternative interpretations.

1) Quirrel said this while he was roleplaying as the defense professor. Perhaps the role he came up with happened to be constructed so that Harry's actions made him happy. But that doesn't mean that it made Riddle happy (insofar as the "true" identity of Riddle exists). Heck, we don't even know that Harry is dealing with Riddle in chapter 108, or yet another identity that Riddle selected for the occasion. (He "changes names as you or I change clothes.")

2) Even if Quirrel/Riddle was not roleplaying, this could have been a calculated statement made in order to gain Harry's trust and admiration. There seems to have been a lot of that going on in Harry's interactions with Quirrel.

Comment by rafiss on A Gamification Of Education: a modest proposal based on the Universal Decimal Classification and RPG skill trees · 2013-07-11T06:15:16.685Z · LW · GW

I'm starting a job in an "adaptive learning" startup soon, and many of the points you make here remind me of the things this company does or plans to do. The basic idea of the company is that it collects data about the student as he or she interacts with an electronic course, then uses this to personalize the course and make recommendations for the student's education path. This isn't quite the same as what you're suggesting, where a student independently finds educational content and then gets certified in those areas. However, there are several similarities.

The tree of learning definitely exists. In your idea, the students find their own paths through this tree, but the adaptive learning way is to recommend paths to each student based on their performance in previous courses.

Both ideas allow each student to have some type of profile that showcases exactly what knowledge they have obtained. Adaptive learning could also add to this profile things that are specific to how the student learns (e.g. is it better to introduce this person this math concept by offering a rigorous proof or by explaining several revealing examples). Something like that could be helpful for the "intellectual OK!Cupid."

You envision a scheme wherein the student acquires access to real or virtual libraries of educational content, with varying degrees of completely independent learning and institutional guided learning. Right now, the company where I will work partners with education content companies (which seem to be mostly textbook publishers), and is only able to offer adaptive learning to students learning through the specific classes created around these textbooks. However, I believe there are plans to make the platform more open, so any course could be built with it.

The main disadvantage of adaptive learning, which your system does better with, is the problem where the teaching and grading is mostly done by the same entity. This seems to be a necessary aspect of adaptive learning, because the data collected about the student as they are taking a test, quiz, or exam is used to learn more about how the student thinks and learns. Still, there is the possibility that an adaptive course can use a test or quiz as a personalized teaching tool, and not only as a way to evaluate performance.

Comment by rafiss on Welcome to Less Wrong! (5th thread, March 2013) · 2013-07-11T05:35:01.312Z · LW · GW

Hi everyone! I've been lurking around here for a few years, but now I want to be more active in the great discussions that often occur on this site. I discovered Less Wrong about 4 years ago, but the Methods of Rationality fanfic brought me here as a more attentive reader. I've read some of the sequences, and found them generally to use clear reasoning to make great points. If nothing else, reading them has definitely made me think very carefully about the way nature operates and how we perceive it.

In fact, this site was my first exposure to cognitive biases, and since then I've had the chance to study them further in college and read about them independently. This has been tremendously useful for me to understand why I and others I know behave the way we do.

I recently graduated college with a major in computer science and a decent exposure to math, having done some small independent research projects in machine learning. I'll soon begin a job as a software engineer at a late-stage startup that brings machine learning to the field of education.

I find that my greatest weakness with online communities is my tendency to return to lurking, even if I find the content very engaging. I hope to avoid that problem here, and at least continue participating in the comment threads.