bramflakes comments on [FINAL CHAPTER] Harry Potter and the Methods of Rationality discussion thread, March 2015, chapter 122 - Less Wrong

6 Post author: Gondolinian 14 March 2015 04:00PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (162)

You are viewing a single comment's thread. Show more comments above.

Comment author: dxu 16 March 2015 09:58:24PM *  1 point [-]

Oh, I agree. HPMoR is entertaining, but it's not good as a didactic work.

That being said, is HPMoR really meant as a didactic work? It seems to me that Eliezer never really intended it to be a vehicle for teaching; see Chapter 1, in which he links to LessWrong at the very top of the page as a way "to learn everything the main character knows". To me, anyone really looking to learn rationality should look at the Sequences, not HPMoR--and from EY's disclaimer in Chapter 1, it looks as though he agrees.

Also, what do you mean by this part?

Harry always just happening to come up with the right answer for no reason

I don't recall that ever happening.

Comment author: bramflakes 17 March 2015 12:15:16AM *  2 points [-]

That being said, is HPMoR really meant as a didactic work?

We are talking about the fanfic where characters routinely block-quote from cogsci textbooks, aren't we?

Comment author: dxu 17 March 2015 01:08:52AM *  0 points [-]

Yes, which could just as easily be an attempt to be faithful to the characters. We know Harry has read quite a bit of cog-sci, and that he likely hasn't quite internalized it to the degree where he can explain the concepts without the use of the terms themselves, and thus resorts to quoting more-or-less verbatim from the material that he's read. That's not to say the novel is intended to teach cog-sci, and in fact there are gaping holes in the cog-sci presented within HPMoR--holes that are not in the Sequences.

Comment author: bramflakes 17 March 2015 10:51:48AM *  2 points [-]

The science material is presented to the reader in good faith, by the protagonist, who is only ever shown to be wrong in his attempts to link the science to magic, not the science itself. If it's attempting to be faithful to the Harry's youthful hubris, then shouldn't there be parts when Hermione says "actually Harry, you've misunderstood Kahneman and Tversky on X, Y and Z ...", like what happens for magical topics?

There is a section on the site called "science" which reads

All science mentioned in Methods is standard science except where otherwise specified (IIRC, the only two uses of nonstandard theories are Barbour’s timeless physics in Ch. 28 and my own timeless decision theory in Ch. 33). Wherever possible, I have mentioned standard terminology inside the book to make Googling easier. At some future point I may compile a complete list for all the scientific references in Methods, but this has not yet been done.

and if that weren't enough, Yudkowsky explicitly states that the science material is meant to be didactic.

Furthermore, "but it's better in the Sequences" is a terrible excuse. How many people are going to read a fun work of fiction, vs a sprawling 888 series of contrarian philosophy essays? A significant fraction of people on this very site have not read them, and then imagine what the odds are for the average fanfic reader (of whom there are an order of magnitude or two more than LessWrong users). Thousands of people are reading this story and taking what Yudkowsky says on faith (did you independently Google every science reference in the story? I sure didn't), so if the science is wrong then that's thousands of people coming away worse-off than when they started, and Yudkowsky is aware of this possibility..

Comment author: TobyBartels 19 March 2015 08:37:08AM 2 points [-]

The ironic thing about those exceptions is that bringing in Barbour's timeless physics is arguably itself one of the errors. In Harry's explanation of how he was able to perform partial transfiguration, there's nothing from Barbour except the phrase ‘timeless physics’; Harry's explication of that, as enforcing a relationship between separate time slices rather than performing a change, is the standard idea of a block universe, going back at least to 1908.