This story brings out newbie defenders like nothing else.

My complaint is about the Gom Jabbar spell name, among other things. I guess you won't blink when Voldemort's horcrux returns to Earth as V'ger, who also happens to be a Vulcan Mentat with a soul-drinking, mithril light saber named Stormbringer and an infinite improbability device. A former Viper pilot, he'll say "Frack" a lot and newbie types will come on here and gush because he said "Bayesian" and therefore must be incredibly clever. The rest of us get tripped up by dropping random crap from other universes into a story. It's like an actor who turns to the audience and winks; hence, the fourth wall comment.

You want a more substantive issue with the story? Fine. For all his talk of Bayesian inference and how important the "methods of rationality" are, Harry shows irrational, unscientific thinking in his "scientific" discussion of the genetics of magic and blood: From a solitary, small-sample-set measurement (without vetting the sampling methodology to determine whether the sample set could have been biased--there are any number of ways this could be so), Harry draws crisp conclusions about a system without couching it properly in probabilistic language.

He then takes the irresponsible step of permitting both he and his minion to go forth with the new "scientific truth" without even considering an obvious alternative: that although the experiment may have been constructed well, it's feasible for one of the alternative hypotheses to be valid yet still express the same outcome within their limited test. How probable is this? It's an elementary application of Bayesian theory to work out. (Incidentally, I've always disliked the term, "Bayesian," since it's just an elementary result from the algebra of conditional probabilities; the mathematical operation of division hardly deserves its own special name in this particular context).

A real scientist would recognize this and express the outcome in terms of uncertainty. If necessary, he would have used the first experiment to guide the design a more conclusive follow-on experiment that would reduce the uncertainty (or disprove the results of the first experiment). He would look at his sampling methodology and consider the possibility of bias. After a series of such experiments, he would have higher confidence that whatever hypothesis is supported by the data is valid, yet would never possess the conceit that he has "proven" anything beyond doubt. Sure, it wouldn't have served the story as well, with Draco's breakdown and the like (the mechanics of doing science is rarely as glamorous as portrayed in fiction), but then it wouldn't pretend to be something it isn't either.

If an author cannot write believably about a character smarter than he (an implicit admission on the author's part that he believes his own brilliance to be on par with Harry's), it could also be the case that one also cannot write believably about the scientific process without being one. This author does not appear to be a scientist.

 

Ah, i think i see the problem. Harry is supposed to be as smart as a 11 years old who study, not the author.

Harry was obviously *supposed* to be stupid there, maybe because he already knew the tests were wrong, but most likely (and said by the author in the AN) because for all Harry's arrogance, he is sadly lacking in practice and knowledge both.

Most of the criticism of the fic i see here boils down to assuming Harry/Quirrel/Draco/AD speak for the author, when they obviously don't.

 

Based on the A/N for today's update, one might be led to wonder whether the author reads this thread.

If so, to the author, I'd say that despite my pessimistic comments in the thread, I do genuinely enjoy this story (as evinced by my still reading it). My complaints are more disappointment in having my hopes unrealized for an epic story, not just one merely decent/borderline good.

 

And such a complaint is complete nonsense. I guess when Rodenberry used a "positronic brain" for his android, he was making a crossover with Asimov, rather than simply making a homage. Now, is this a prediction of yours extrapolating from a small-sample measurement or just a commitment to interpreting everything in the worst possible light? It's not as if this story has no actual faults. You don't have to make up nonsensical ones. Really.

You're the only person to even mention this word in this thread until now, but what's much more ironic is that it's quite clear that you've no understanding of what it means. Perhaps you should take a breath once in a while while playing on your bandwagon.

Yes. Thanks.

Of course. The author even outlined some of the mistakes Harry made in the author's notes. He's an eleven-year-old that happens to rabidly committed to the ideal of rationality, not an oracular incarnation of such.

Umm... no. It's a semi-independent theory of statistics, in that it differs with the orthodox statistics in philosophical interpretation of probability (long-term tendency or reasonability of belief), methodological focus (repeatability of controlled experiments or inference), and in some extreme cases contrary conclusions. While I think sometimes the differences are exaggerated, substantial ones do exist. Somewhat relevant here, a Bayesian would coach his conclusions about hypotheses in terms of probability, rather than confidence (in the technical sense of more orthodox statistics).

That's a good point. Part of it is already hinted in the story itself--Harry requested a sample size of forty, and Draco quite tellingly fell short of that. Whether Draco's sample is good enough, particularly since it already has one flaw is an obvious issue, and 'why forty in the first place and why here' would be a good follow-up one.

Plot-wise, what was important is to get Draco to admit that his worldview even could be falsified. I doubt a full analysis would serve the story any better, but still, some more doubt on his part would have been very appropriate.

I'm sorry, what? Harry has a conclusion, and your criticism of his failure to coach it in probabilistic terms is both correct and very relevant, but in what crazy world does this mean that he considers it "beyond doubt"?

It's an explicit admission on the author's part that the intent is for Harry to make the kinds of mistakes that the author would have made when he was eleven. Whether or not this is still overly generous indicator of the author's intelligence is of course very debatable, but also irrelevant to the issue at hand. Whether or not the story would have been better off is Harry really was an unerringly correct supergenius is a more relevant issue, but I think the answer to that is "hell no."

P.S. I'm actually pretty annoyed at the author's grinding of the Bayesian axe in regards to some certain scientific theories, but that has more to do with his blog than the story.

 

He could have called the spell anything whatsoever. That he chose to have a pureblood borrow a device from the novel, Dune, to name his spell is distracting "wink, wink at the reader" derailment to me. I guess you'd say that I'm "making up nonsense" when I say that I and others find this practice distracting.

I rarely read author's notes; a story should stand on its own without having to be explained by the author. Interestingly, the few I have read often appear to riff off of what is written in this thread (including the latest A/N, which addresses my earlier comment). More on the substance of your comment later.

Bayes's theorem itself is simple division, but you have a point for the rest. My delving into Bayesian inference is limited to works by Jaynes and a few others; I'm far from a statistics expert, though I know a bit about the field as a professional scientist. I too expected more attention to the probabilistic interpretation of the experiment, given how frequently "Bayesian" comes up in the story (the name of their cabal has Bayesian in the title, after all).

The questions of sampler bias and adequate sample size should have at least been stated, even if just in Harry's head. Harry very clearly intended for Draco to not grok the probabilistic nature of the measurement, but rather to draw the conclusion of crisp, 100% certitude. Perhaps he did it because he had an ulterior motive (having Draco reject blood purity and accept that his ideas may be falsifiable, as you say) that trumps adherence to the principles of rational inference. If so, he's a hypocrite, permitting Draco to believe something that's wrong because it's more convenient immediately after lecturing him against such practice.

Did Draco go to paintings representative of all houses, or just those he felt comfortable with, famous pureblood lines, say? Might the groups willing to speak with a random Slytherin student be less forthcoming with their data on Squib relatives? How exactly did he ask the question--could it have been phrased in a way that biased the results? These lines were not pursued for a number of reasons, I'm sure, but knowing that they were at least considered would have been worthwhile and somewhat less distracting to me.

I could find no instance in this or the last chapter where Harry himself expressed doubt over his conclusion after having been told the data. Yes, I agree that it's the most likely interpretation of the data, but if he were a card-carrying Bayesian, he'd carry around that little sliver of doubt (and quantify it, if possible). Without, the evidence supports that his reasoning falls into the same fallacy as the game he played with Hermione on the train with the ascending numbers.

Fine, fine, but it's also a convenient dodge whenever Harry misapplies stuff: Set up strawmen, claim credit for the ballyhooed "methods of rationality" when it suits the author, then duck behind "well, he's just a kid" when problems are identified with Harry's logic. The kid explanation always seems to come after the fact too.

Does this story say much, really, about the methods of rationality when we're shown continually that Harry is incompetent in applying them?

 

You guys are way too fucking TLDR.

I'm with pers on this one, tbh. He's not saying he hates the story and it's fucking terrible, he's just pointing out some legitimate concerns.

As for the new chapter? Haven't read it yet but....

I haven't really seen any proof of the authors arrogance until now. Christ... Either way, off to see how the new chapter is.

 

He does. An AN that I can't find back (perhaps I'm not looking at the right chapter or perhaps it has just been changed) mentioned DLP's reaction to the fic. Something about how the fic was taking off and even DLP was starting to warm up to it...

 

Just read the chapter, wasn't great, but it was no where near as bad as the last 3 chapters, thank god.

 

WTF? No offence Pers, but that's ridiculous. By that definition, Ben Afleck can cipher with the best MIT professors. You have to consider your audience. Sure, if the person watching Good Will Hunting happens to be a mathematical savant, they may find a good deal to critique about the realism of the character. For the rest of us, it was convincing. I'm hoping I took your comment out of context as I hastened to skip the ridiculously long-winded scientific diatribe. If there is anything I want to read less than a lengthy scientific lecture in a fictional story, it's a lengthy critique of the lecture.

Let's just agree than Harry in this story is not as clever as he thinks he is, and neither is the author. Neither am I, for that matter, but that's beside the point.

 

Leaving aside the scientific details (I'm about as far from being a scientist as it's possible to be), I'm still finding this incredibly frustrating. I read chapters 6-17 earlier, and the humour often comes across as forced. Some of the characterization is feeble - I really don't like his Dumbledore so far, for instance. And while I've no idea about the accuracy of the science stuff, as said, it gets incredibly tedious at times. Whatever the merits of the content or idea, the execution is lacking, at least for me.

But all that said, there are genuinely good moments. I like his take on Quirrell, and the Defence lesson where they find out who the most dangerous one among them is was impressive. When Less Wrong just lets the humour flow, instead of forcing a ridiculously elaborate joke, it's quite amusing - although I did like the whole Time-Turner chapter, basically pointless as it was.

So yeah. Technically sound, largely frustrating and the occasional flash of quality. I might pick it up again when I'm bored, but it's never going to be more than that.

 

Out of context, this does sound asshatty. (Hell, in context, it's a little snarky).

To clarify, I borrowed from the text: if you read the second-to-last chapter, Lucius said to Draco that those who write plays can't possibly write clever people effectively because they aren't as clever as their subjects (something that has a meta-meaning for the author himself). I advanced the conjecture as a throwaway comment at the end of my post that one might extend this to the practice of doing science.

As an aside, it's hard for a professional physicist to find readable science fiction because it's ridiculously easy to spot stuff that just ain't right. I've been told by my lawyer friends that reading John Grisham novels is the same for them.

 

That's not at all what Lucius said, which was a much more narrow statement (and even that was debatable).

 

"Much more narrow" is itself debatable.

Welcome to DLP.

 

Scientific validity aside (I must admit I skimmed most of the "heavy" science mentioned), I enjoy his characterization. Quirrel is interesting, and unlike most others, I enjoy his Dumbledore - especially in the most recent chapter.

Honestly, he probably writes most of the secondary characters a lot better than he does Harry.

 

Gotcha. Well, that's the problem and the solution Pers. Don't pay attention to the author's views outside the story. But, I can see what you are saying. Because of the lecturing (preaching) in the story, we are thrust into the author's views. I'm only able to continue being a fan of this story because I skim those bits and view this as a crack!fic. The author may not, but I do.

That just plain sucks for you. Too bad you can't just shut it off. There's an awful lot of good sci-fi that is probably more full of holes than my underwear drawer.

 

No, but I would say it's making a mountain out of a molehill (though I may be conflating your statements with Yume's earlier ones, in which case sorry about that).

It does stand on its own at least in this particular respect. Projecting intended perfection onto Harry is unwarranted regardless of whether or not it was disclaimed in the author's notes. You're reading a story about an eleven-year-old that's obsessed with reading and scientific inquiry. Why would you assume that he will do it better or even on par with people who've actually had formal training in it, unless he actually demonstrates such competence in the story?

You wrote (emphasis mine):

Those are two very different things. Perhaps you meant to say 'theorem' above, but in any case your question as to why 'Bayesian' has is such an emphasized label, then that's why: the the referent here is not the theorem about conditional probabilities, but rather an entire school of statistics and interpretation of probability. The details of the differences aren't really important here (but alluded to in the previous post).

You're absolutely right about that--the shades of hypocrisy especially so.

I don't think that's warranted. First, he's nowhere that consistent about getting it wrong. But more to the point, do you really see this fic as primarily a lecture or tutorial on how to reason, rathter than a story about someone with deep commitments to rationality trying to make sense of magic?

I think of this story at least 'should be' the latter--if he takes on the former, he's bound to fail at telling a story--which is why I'm disappointed when the author goes into explicit long-lecture-mode (which is a bit too often... though the shorter tidbits are frequently pretty interesting, and sometimes he manages to plausibly insert a deconstruction of why something's wrong without obvious heavy-handedness). That nitty-gritty is better off remaining on his blog... or (le gasp!) actual textbooks.

 

Darn. It's no fun when we find we're mostly in agreement, disagreeing only on the less vital stuff. Thanks for the discussion, Vorpal. (I'll be on travel tomorrow and won't be back on to reply to anything for a day or so).

I stand by what I said before--this is an entertaining story. I have a short list of things I wish the author had done vs. what he did do, but in my advanced age (ancient, in fact--I'm nearly 40, about to become ineligible for a Fields Medal!), I've learned to live with a bit of disappointment.

 

You know what I do as soon as I watch an episode of House MD? Read a real doctor's list of mistakes for that chapter.

 

New chapter...

Better than the last ones. Still not that good.

I'm starting to think that we've seen the highpoint of this fic, but who knows what could happen. The plotline seems to be all over the place too be honest, I thought we were going to get some excellent development with Harry getting given Riddle's diary, but apparently not.