Monday, January 31, 2011

Thoughts on Frindle

Frindle
Andrew Clements


Just from the title, I have to wonder if it had any connection to the Kindle.  A quick search reveals... that you can buy Frindle for the Kindle, but not if Frindle helped inspire the Kindle name.  Maybe not.

Anyway this is a kids' book from 1996; I'm a little amazed I'd never heard of it.  The main ideas about dictionaries being descriptive rather than prescriptive, on a fundamental level, and the general nature of language as a social construction, are some cool ideas that I kept encountering through high school and beyond.  I like that this book presents those ideas in a very accessible form.

I identified with the use of dictionaries in school.  I remember, vaguely, doing vocabulary lists and looking stuff up in paper dictionaries when I was in grade school.  But I have to wonder, do kids do that these days?  Is anybody still really using paper dictionaries?  Even when I was in grade school, all those years ago, the class set of paper dictionaries was getting to be a bit dusty and unappreciated.  I also remember using the thesaurus in Word a lot.  And now I go to dictionary.com or dictionary.google.com exclusively.  The internet seems somehow inherently more flexible, more the creation of human hands, than a musty hard-bound dictionary.  Anyway.

One disappointing thing: the author presents a story about the origin of the word "quiz" as if it were factual.  (Well, it comes from the mouth of a TV announcer, so maybe we're supposed to acknowledge that they'll say anything that sounds good.)  Anyway, the claim is that "quiz" was invented by one person, the way "frindle" is in the story.  I checked wiki, and according to wiki the story is apocryphal.  So there's that.

Language seems to spontaneously reject deliberately coined words, like "frindle" in this book.  It's like people don't trust their authenticity, if somebody's consciously trying to make them.  Doesn't seem real.  I can't think of an example of somebody making a word just for the heck of it, like the kid does in this story.  "Quiz" is out.

The story's logic is consistent with my earlier thoughts about the artificiality of the war between descriptive and prescriptive linguistics.  Parents (teachers) teach their children language, and they teach it the way they learned it.  If they didn't, language would change far too rapidly, with wild creoling every generation or so.  And that's my description of how prescriptivism is an essential part of language.

Anyway, fun book, reads in about an hour or an hour ten.

Tuesday, January 25, 2011

Thoughts on The Grand Design

The Grand Design
Stephen Hawking and Leonard Mlodinow


I bought a copy with a Christmas Barnes and Noble gift gard from my grandparents, but I just started working on my philosophy again and I got to a certain point and realized I wanted to read this book.

So I went home and grabbed my copy and started reading, and that was yesterday.  This thing reads really fast.  It's clearly written to be read easily, by anyone.  There are even jokes.  I don't know whether Hawking or Mlodinow is chiefly responsible.  But it's an easy read.  It might help that I learned about a lot of this stuff in college or thereabouts.  I actually did a version of the double slit experiment in a college physics class.  Although I guess that was just with a laser, and who do you need to convince that light "is" a wave?  Anyway...  I think pretty much anybody could read it.


They go through a sort of history of science, and then get to their epistemology, or ontology, whatever, anyway they call it "model-dependent realism".  Not quite realism ("I believe in an external world") or anti-realism ("the only thing I believe in our my thoughts")*.  This is the first interesting thing in the book.  Unfortunately I don't think they really explain it very well, instead kind of using it to explain things away that I don't think it really does.  I want to look into it more, but the rule is first write my thoughts, THEN check the wiki.  Also, it seems like it could be related to the thesis of another book of mine that I should really get around to reading, Relativism and the Foundations of Philosophy, the thesis of which is something like "everything true is relatively true".  On the face of it, this does not seem satisfying.  I really want some things to be absolutely true, damn it.

Also, and sort of related-slash-segueing-to-the-next-thing, the book really reminds me that scientists believe some wild things these days.  From the glossary:

Classical physics: any theory of physics in which the universe is assumed to have a single, well-defined history

And that's been out since the 1920s or something!  The religion of "I believe in physical things, like chairs and atoms" cannot be called the religion of science any more.


Interesting thing number two is my second-biggest gripe with the book, which is that quantum-mechanical "observers" are not adequately explained.  They make something of an attempt, saying that observing changes things because (for example) the photon that you see hit the particle you're seeing.  But damn it, a bunch of photons "hit" the particle without you seeing them, and that doesn't cause any wave functions to collapse!  I HAVE A BIG PROBLEM WITH THE CONCEPT OF OBSERVERS IN QUANTUM MECHANICS SOMEBODY GIVE ME A SATISFYING EXPLANATION PLZ


Interesting thing number three is that in the final chapter, all of a sudden they're writing mostly about Conway's Game of Life!  Holy crap, I haven't seen that much coverage of the Game of Life since I was looking into it as a possibility for making an artificial universe simulation - which is uncannily similar to how the authors use it in the book.  They seem to be implying that the universe (our universe) may actually have some simple (or simple-ish) laws, that we will never know - or maybe even if we knew them, it wouldn't be useful.  They stick to their idea of "apparent laws" (although I don't like how they define this term in the glossary; it seems different from how they use it in the book - and also they have a bad habit of saying that laws "govern" the universe; I think that language should be avoided, especially since with "model-dependent realism" the emphasis is on modeling, not specifying.  I do think it's cool that GoL is in there; it made me think of Wolfram and his New Kind of Science (which I haven't actually read).

I had something else to add here...

Oh, maybe it was what they said about free will.  Basically they said that it doesn't exist, but that it should be taken as an "effective theory" because we can't run the math to make useful predictions.  To me, this seems to be an incomplete or at least unsatisfactory treatment of the issue.


Okay so there are 3.5 interesting things, and here's number 4.5: my biggest gripe about the book:

In the first chapter they claim that they will address these questions:

Why is there something rather than nothing?
Why do we exist?
Why this particular set of laws and not some other?

Okay let's say the first two questions are the same question and deal with them last.  Question three, then, they answer in two ways.  Principally, using the strong anthropic principle.  They say this is okay because their other theory predicts a lot of universes.  Okay that kind of makes sense.  But their other theory is M-theory, which, according to their description, is a lot more like a pile of paint chips than the Mona Lisa.  They never say exactly what it says, it's just a collection of theories, and generally it does not impress.  So the reason why M-theory and not some other is that it predicts some 10^500 possible universes, and boom we're in a universe, so hey didn't everything work out well.  I'm going to take some persuading before I get on board with M-theory.

But that's not even the big gripe!  Questions one and two!  You have to pay attention in the final pages to catch it, but the answer is this!

"Spontaneous creation is the reason there is something rather than nothing, why the universe exists, why we exist." (p.180)

WHOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

Crappy answer.  "The universe exists because it just does!"  Well why does it?  "Well we've got this M-theory, and M-theory says this kind of thing happens all the time.  Why, you just won't believe how many universes sprang into existence just this morning!"  Well, why is there M-theory?

At that point you get either a) an appropriate screeching halt, as you start to realize that you can't ever beat the kid asking "why?", or b) a return to the model-dependent realism, that M-theory matches our observations of the universe, but that's circular, or c) I really thought I had a third option but now it's escaped me.


I really liked the book, in that it was interesting and fun and it seemed sort of like Hawking and that other guy were trying to do the same thing I'm trying to do: resolve philosophy so everybody can get back to watching the super bowl or whatever.  But I don't think they've done it.  And maybe M-theory is a great theory, but I haven't seen any evidence of that yet, really.  Of course it's probably hard to put into a mass-market book, but still.  Maybe I'll check the wiki.


Post-write:
Read this review by Penrose, which served to remind me that not all scientists agree with Hawking, and also showed me that I'm not the only guy skeptical of M-theory, and model-dependent realism.  (Penrose seems to be a pretty firm realist.)

Crazy review by this German guy.  Wow, I haven't read anything with so much sarcasm (or is it irony?) in a long time!  People still write this way?  Is it just a German thing?  Anyway, he is not impressed at all by The Grand Design, which is making me feel more comfortable with my doubts as well.

Oh and there's no wiki for "model-dependent realism" - I guess it's just Hawking's term.  And nobody seems to take M-theory very seriously, so I don't feel like looking into it much.  A sort of string theory.

And I saw something on reddit for the first time.  I don't think that's the real Stephen Hawking there, on reddit.

And... one more: some guy who seems to be in agreement with Hawking.  But I'm still not satisfied with dark matter or dark energy, because they seem to be big hacks.  And how did you measure the total energy of the universe, guy?  How?  Anyway, if the total energy of the universe is zero, I don't see how that makes it self-starting, necessarily.  It doesn't explain why there should be rules that allow for self-starting.

Tuesday, January 18, 2011

Thoughts on The Year of the Flood

The Year of the Flood
Margaret Atwood


I read this sequel to Oryx and Crake because, according to the wiki, it was supposed to tell me more about Oryx, the mysterious female lead, and also what happened after the cliffhanger ending.

In the first, I was disappointed, because there's hardly any mention of Oryx, and her origin and motivations are pretty much as opaque as before.

In the second, the cliffhanger was "resolved" and then the book ended just seconds later, with another, weirder cliffhanger.  Whoo hoo.


Continuing to compare everything I read to Vonnegut's stuff, if Oryx and Crake was like Galapagos, The Year of the Flood is like Cat's Cradle.  It focuses on a quasi-religious group called The God's Gardeners.  They talk about god a lot but don't necessarily all really believe in it.  The book is fairly ambiguous about whether the religion is really helpful to its members.

In some way I guess my gripe with the book is that it's too realistic, where I would rather find some moral or something.  Everything is just so pointless, all through the book.  It's like real life, I guess, but it isn't satisfying.

Of course, there are ridiculous coincidences of meeting and so on, but whatever.

The writing is technically a little disappointing.  In the first chapter or so I felt let down a little by the writing.  It could be that it's supposed to be, in places, the writing/speeches/songs of The God's Gardeners, who are only human, after all, and a little crazy.  So if the lyrics of the song are bad, you could just say hey, it's because the God's Gardeners are bad song-writers.

The most fun thing about the book, actually, is that somebody took the lyrics from the song and made them into full-fledged songs.  They're much more believable being sung.  I bought them all online and enjoy them in a weird way.  I usually can't stand religious music because it makes me sick to think that people are walking around believing in the tripe.  But The Hymns of the God's Gardeners are made up and slightly silly in places, so I can listen to them in a much more detached, fake-ethnographic sense, I guess.  Fun.


I still think it's pretty dumb having the whole book be about how awful it is what people do to the earth, and then in the end it's just one crazy guy who deliberately kills everybody.  I guess I could interpret away my objection.  Whatever.

Saturday, January 15, 2011

Thoughts on a computer playing Jeopardy

IBM made a computer that can play Jeopardy really well, as reported here.


On the one hand, of course a computer should be great at trivia.  It can remember everything you put in it, perfectly.  Computers are really good at remembering trivia.

On the other hand, it's impressive that they've got the language processing working well enough to successfully match the questions to the answers.  Language processing is hard for computers.

But even if this computer can beat people at Jeopardy, as computers have for some time beat people at chess, I'm not sure I should be that impressed.  The computers are definitely achieving ends that humans achieve, but the means are so clearly contrived and mechanical.  The computers don't do anything that I would call thinking.


There is a divide, I guess, in AI:

A: Build a system by whatever means, usually explicitly oriented toward an outcome, or goal performance.  The Jeopardy computer does this, although it does do it by a pretty neat unstructured data analysis method using Apache, Java, and C++, of all things.  Chess computers do this.  This approach is usually pretty quick to achieve some results, and can scale to really good results in specific applications (chess, Jeopardy).  But it doesn't transfer to other domains.  I can't deny the usefulness of this approach, but it's just not very interesting to me.  It's like house painting.  It gets a job done.

B: Build a general system, possibly based on human biology or general ideas about thinking (neural networks, etc.).  This kind of system has a harder time excelling at specific tasks like playing chess, but has more promise of future flexibility, and of becoming a kind of intelligence comparable to what humans have.  It may be a long shot, but I think this stuff is more interesting, and I think it is more likely to be really cool in the future.


So, like the chat program that fooled people into thinking it was human by deliberately making spelling mistakes, things that are explicitly programmed by humans may achieve short-term success in special case situations, but are not the right path to real advances in AI.

Monday, January 3, 2011

Thoughts on Oryx and Crake

Oryx and Crake
Margaret Atwood

I don't remember exactly how, but somehow this book was recommended to me by some combination of Toni, Brice, and Joanna, I think. Maybe others.

So I borrowed the book and read it on the way back from Wisconsin to Korea. First of all, I was surprised how quickly it went. Reading in English is so easy, and novels are so much fun to read! So I did the whole thing in probably around ten hours. Maybe I should read more books like that.

Atwood is compared to Orwell on the cover, and I guess it is a little like 1984 with the totalitarian government replaced by competing corporations. I don't know if I would have made the comparison if it wasn't on the cover like that.

The comparison I made myself is with Vonnegut's Galapagos. In Galapagos an arbitrary (implied natural) microorganism makes the entire population of humans, except for a handful stranded on a remote island, all turn sterile. So humanity is done, except for those isolated humans who go on to evolve into weird seal-like creatures.

Sort of the same thing happens in Oryx and Crake. Well, Crake starts to make everybody sterile, but then immediately afterward just kills everybody with an artificially manufactured microorganism. Anyway all the humans die, except for one and possibly a handful more, and some genetically engineered neo-humans, as a sort of new Adam and Eve population.

Note: one interesting thing is that new population is supposed to be designed to have none of the problems that people have, but toward the end there are signs that some of the problems are reemerging among them anyway.

So in both Galapagos and Oryx and Crake, pretty much everybody dies, with just a few, pretty much non-human, survivors. In Oryx and Crake, it's somebody's fault.

A complaint I often have with stories is that there's some sort of pure evil force - with no sensible motivation. Happens all the time. "I want to kill everyone! Destroy the world! Etc.! Because I just do." So a lot of the book is about this Crake guy, a super-genius who kills all humans. The way it's presented, it seems like it's mostly because he's pissed off about the death of his dad and therefore disgusted with humans. Either that, or he loves humans but things are getting so shitty on the planet because of over-consumption etc. that he thinks it would be better to kill everybody off quickly; sort of species euthanasia.

It's not totally clear (or maybe I'm stupid) and the ending is also open. I'm not sure how I feel about open endings. I kind of think I dislike them. It's like the author is assigning homework at the end of the book instead of finishing it. "What do YOU think happens?" "Hey lady, it's YOUR book!"

Anyway I think Snowman kills them and then dies himself shortly thereafter because of his infected foot.

Other things: the book is largely about the evils of over-consumption, pollution, capitalism generally, and genetically engineering everything. But at the end the huge crisis and death of humanity is not the fault of any of those really, but one guy that just wants to kill all humans.

Anyway, final opinion, it's not a bad story, but not quite as great as some of the blurbs make it out to be. Sometimes there's great subtlety and dark humor, but sometimes it's too obvious and smug about it.

And I'm not going to go look up a German quote thing to find out what it means. I did feel smart when I knew the Latin and French bits though, thanks Marge.


Reading the wiki, I see that I missed an unattributed quote from Slaughterhouse Five that's in there! So there is some explicit Vonnegut. Vonnegut rocks.

Also, apparently Atwood DOES finish her story - in a 2009 sequel called The Year of the Flood. I should probably go read that now.

Thoughts on written language

Rambled on the plane from Chicago to Japan, 2011.1.1-2.


Written Language
or
Yes, I will tell you what I think about typography, or some related issues, at least

Let's take graphic novels on one side, and plain ASCII text on the other. ASCII, if you don't know, is a standard computer thing that allows you to record lower and upper-case English letters, along with the standard punctuation. Nothing fancier than a hyphen and underscore, not even smart quotes. Just plain text, as it is sometimes called. Graphic novels, if you don't know, are comic books with aspirations.

The point is, there are differing levels of graphics magic that can go into things that fall roughly under the heading of "the written word".

In plain text you can't underline, or put things in italics, or bold. It's really very restrictive. Just the words. You can create some sort of emphasis-alternative by putting things in all caps. In my opinion this is almost never the right thing to do.

But normal books these days do all of the italicizing and bolding and underlining they want. I notice it but usually it doesn't bother me. It does lead to annoying things like having to mention whether it was the author's emphasis or not when you quote it or whatever, but okay.

And then also a lot of ostensibly normal straight-up books will interrupt the text to put in a whole picture or something. My beloved Kurt Vonnegut does this (rather a lot) in Breakfast of Champions. He writes things like, "He ate an apple. Apples look like this:"

And the next thing in the "text" is a line drawing of an apple. Even Harry Potter has some stuff like this; little sketches inserted here and there - and I don't mean the decorative illustrations; the little pictures form a part of the "text".

My gut feeling is to be opposed to this. I am not quite sure exactly why, or where to draw the line. Graphic novels are their own thing, but in things that are otherwise just text, novels and the like, I think a line needs to be drawn.

In the case of pictures, it seems to violate the idea of the text recording speech, just telling a story. You don't think of telling a story around the camp fire, and then stopping to hold up a picture. I guess you could stop and draw a picture in the dirt or something, but it's really different from just TELLING the story.*

There's also issues with pictures, like the difficulty "quoting" them, or changing them to audio books or Braille, also translations if they contain text, etc.

I am pretty sure I can unilaterally condemn pictures in novels. Novels are not picture books. They are for reading. That's why graphic novels have a different name, and one that might be better if it was even more different.

Also, as a person who is literate, I am used to writing without sticking pictures in everywhere. It would be annoying and a little difficult to add a picture to this, as I type it out with my thumbs.

Is it just because of convention, because of habit that I don't want to add pictures? Will people in the future add a lot of pictures all the time, with big touch screens everywhere and appropriate software for easy drawing?

But pen and paper has been along for longer, and you aren't considered a good author if you throw up your hands and say "Heck, it's hard to describe but it looked like this:" and then sketch it out. I think the key reason that pictures are usually kept separate from the written word is that the written word is modeling the spoken word, which is necessarily non-graphic.

So what should be the building blocks of "the written word"? Just 26 letters and the space? Lower and upper case? Punctuation? Italics? Bold? Underline? Different font sizes? Different fonts? Different font colors? Smileys?

Would Shakespeare be better if he'd used fonts to convey some of the players' emotion?

As anyone who's texted, chatted, etc. can tell you, it can be difficult to convey your message in ASCII. Hence all the extra cues people use.

2b r not 2b lol
THAT is question j/k

That sicks me out.

Maybe only my family used that expression. As you can guess, I'm not a fan of all that in published writing. I don't want to read a book of it. Sure I use it for communicating with friends.

Here I think the problem really is that written language does not actually capture spoken language as well as we would like it to, or as well as we would like to pretend it does. In speech we rely on shortcuts like inflection to pack more meaning into words. I try to put some of that into things I write too; I left in the all-caps I instinctively resorted to above (at the *).

I think the very real disconnect between spoken and written language is something that should be more clearly acknowledged, perhaps. I think of Latin, which served as a sort of written-only (almost) language for a long time in Europe. Being dead might not be so bad for a language, if your goal is just clear communication of ideas between various people.

As a note on punctuation, I think what I did above with the asterisk and most recently with parentheses are kind of cheats. I could probably improve that with editing.

Also in Korea for a long time there was no written form of the Korean language(s) and the elite used Chinese in a role roughly similar to that of the aforementioned Latin. But in that case I am forced to think of the way in which Chinese characters were used to write Korean poems in transliteration. Apparently at least some Korean people were not so interested in doing their poetry in Chinese.

Maybe it's appropriate for writing that is trying to capture the human voice, like poetry might, like dialogue, like a story can, should make full use of whatever technology is available. Maybe a novel can be a better novel for having a bunch of typographical trickery. But in that case it makes me think that we need some kind of though-out genre distinctions. A novel is a novel and a graphic novel is a graphic novel; are there a thousand grades in between?

If I look at standard practice, it seems like italics is pretty popular. I guess it's mostly replaced underlining, maybe because it looks better and people don't rely on type-writers any more. Bold is pretty popular too. Now that I think about it, underlining seems to have gotten really rare.

In books and newspapers aside from headlines, writers almost never change font or font size, probably because it's just so obnoxious - both to change and to read.

So maybe I should accept bold and italics as part of the standard set. But I really don't want to. Is it just romantic attachment to ASCII? Is it that it won't copy and paste cleanly on ever web form?

What would I do if I was writing a novel? I'm writing this whole thing because of the italics in Oryx and Crake. I didn't really mind reading it that way.

But in the end I don't think I could be happy with it for myself.

But then you have stuff like MLA format demanding all these typographical things, argh... And even as I bemoan that I can't get away from using obviously vocal things in my writing. Is that just laziness?

Is this all heading for a conclusion other than "written language, like spoken language, is guided by social conventions specific to audience and application, and these conventions can change with time"?

Maybe not. But I don't like all that darn formatting sneaking in everywhere.