Re: Olha só que idéia interessante do Phil Agre

[Ricardo Ueda Karpischek <ueda@ime.usp.br>]

Faz algum tempo eu procurei freewares para aprendizado de
línguas, mas não encontrei nenhum. Existem alguns softwares
comerciais para aprendizado de línguas, já cheguei a testar
dois.

A idéia do Phil Agre poderia talvez ser implementada na forma
de um sistema de avaliação. A pessoa interessada
faria um teste via web ou email, e a partir dos seus acertos
ou erros o sistema analisaria quais palavras essa pessoa
conhece ou desconhece, como naquele sistema do Daniel Menascé.

Eu tenho dois testes online muito simples, um de TCP/IP e
outro de Unix:

    http://www.ime.usp.br/~ueda/tcpaval
    http://www.ime.usp.br/~ueda/uxaval

Entretanto eles não fazem accounting do que cada interessado
sabe ou não sabe.

O extender do Phil Agre tem também algo a ver com a própria
atividade típica do dicionarista. O dicionarista coleta das
fontes escritas novos vocábulos, e isso à vezes é até origem de
polêmicas: "Desafiado por um sceptico, certa vez, a descobrir
tres lacunas nas edições de tres grandes jornaes brasileiros
publicados num mesmo dia, demonstrei em publico que as
tiragens de dous jornaes paulistas e um carioca, de 17 de
fevereiro de 1924, me davam 106 lacunas do diccionario do
Snr. Candido de Figueiredo!" (Affonso Taunay, Reparos ao
Novo Diccionario de Candido de Figueiredo, 1926).

Um abraço,

Ricardo Ueda.


On Wed, 1 Nov 2000, Imre Simon wrote:

> Date: Sun, 29 Oct 2000 06:15:25 -0800
> From: Phil Agre <pagre@alpha.oac.ucla.edu>
> To: "Red Rock Eater News Service" <rre@lists.gseis.ucla.edu>
> Subject: [RRE]notes and recommendations
> 
> ...
> 
> **
> 
> Wish list.  I want an automatic vocabulary extender.  This program
> would analyze the writing of many thousands or millions of people,
> at their request of course, to determine everyone's vocabulary.
> Then it would compare people's vocabulary lists and suggest the
> next words that each person might want to add to their repertoire.
> So, for example, if I use a certain list of words and you use all
> of those same words plus a few more, then I might get a message with
> definitions and usage examples of those extra words.  Dictionaries
> would be annotated to distinguish between words that are broadly
> useful -- such as the hundreds of Latin adjectives and verbs --
> and those that are more specialized -- such as botanical terms.
> If you and I have the same interests except that you also write about
> plants, then it wouldn't make sense to recommend that I use the word
> "rhododendron".  I probably know what "rhododendron" means already,
> but because I am uninterested in plants I never have any occasion
> to use it.  But I might be interested to know that I have never used
> the words "conflate" or "ingenuous".  It probably wouldn't take more
> than a half-dozen markers on each word to determine which vocabulary
> differences between people are worth flagging.  A more advanced
> analysis might look up the thesaurus entry for words that I overuse
> relative to people with otherwise similar vocabulary profiles, but
> that would be a lot harder, since many of those words are keywords
> in particular intellectual, political, and cultural traditions, for
> which synonyms won't do.
> 
> **
>