[Haskell-cafe] Re: Is Haskell a Good Choice for Web Applications? (ANN: Vocabulink)

[wren ng thornton]

Chris Forno (jekor) wrote:
> The idea is that I spent years studying different languages, generally
> with a textbook. The textbooks tend to focus on teaching rules and
> grammar, with a little bit of vocabulary and dialog each chapter. I
> think the focus should be reversed.

This varies wildly by textbook, with some bias for the language being 
taught. Personally I've found too many vocabulary textbooks and  far too 
few grammar textbooks (that is, actual *grammar* textbooks not 
sentence-sized-vocabulary textbooks).


> Obviously grammar is very important. But is reading about it effective
> for everyone? 

In my experience learning and teaching languages, this too varies wildly 
by learner. Some people do better with an "examples first" or 
"vocabulary based" style where they must come to an intuition of the 
grammar rules; other people (such as myself) do better with a "rules 
first" or "grammar based" style where they must come to learn vocabulary 
on their own.

Neither variety of person is superior nor, as far as I can tell, more 
common at large; so any good teacher or textbook should balance these 
"bottom up" and "top down" approaches. IMO vocabulary is easy to learn, 
it just takes time, whereas grammar is harder to figure out on one's 
own, and so is the better thing for a teacher to focus on. However, this 
says little about reference material (as opposed to learning material), 
and study guides walk a line between reference and teaching.

JGram <http://jgram.org/pages/viewList.php> is an interesting study 
guide that takes a middle path, treating syntactic patterns the same as 
it does lexemes. This is particularly appropriate for a language like 
Japanese where it's not always immediately apparent whether something 
belongs to the "grammar" vs the "lexicon".


> The only way I
> successfully became fluent in a language (Esperanto) was through
> immersion,

This is, hands down, the best way to learn any language. For it to work, 
as you say, some vocabulary is necessary; however, I think the amount of 
vocabulary needed at first is not so large as some think. Daily 
small-talk for getting/giving directions, ordering food, and the like 
comprise a large portion of beginner's language and requires remarkably 
little breadth of vocabulary (a couple hundred words or so). Small-talk 
also includes some of the most obscure and difficult-to-master 
grammatical patterns like greetings, getting the right tone of 
politeness/familiarity, and knowing what sorts of sentence fragments and 
other "ungrammatical" patterns are perfectly acceptable.


> And of course it has very forgiving sentence and a rather simple
> grammar, but I'm finding the experience to be very similar with Japanese
> so far.
> 
> That being said, Esperanto, and even Japanese sentence structure perhaps
> is not as different as an agglutinative language like German. I'll need
> to study it more to find out.

Actually, Japanese is agglutinative too (moreso than German is). The 
basic structures of Japanese are quite simple, however the details 
needed for fluency are quite intricate. Phrase order is rather free, 
though it is not entirely free and it is easy to reorder things so that 
they no longer make sense to native speakers. Aside from a few of the 
common mistakes beginners make, if you mess up the cases/particles 
you'll end up with gibberish.

-- 
Live well,
~wren