Re: ducktyping

[David Given <dg@...>]

Veli-Pekka Tätilä wrote:
[...]
> I can think of two problems with duck typing, which I run into sometimes:
> 
> 1. Naming.
[...]
> 2. Sometimes the exact interface is not formally specified even in the
> docs.

Duck typing's been around for years, although the name and the
trendiness is quite recent. Like most hip new computing technologies,
Smalltalk had it in 1980; it solved the above two problems by requiring
(well-written programs) to specify all their interfaces (although it
called them protocols). These had nothing to do with the actual *code*,
you understand, they're just an organisational thing.

Here's a good and short read:

http://users.ipa.net/~dwighth/smalltalk/byte_aug81/design_principles_behind_smalltalk.html

Notice how familiar a lot of it sounds. Then look at the timestamp...

-- 
┌─── ｄｇ＠ｃｏｗｌａｒｋ．ｃｏｍ ───── http://www.cowlark.com ─────
│
│ "All power corrupts, but we need electricity." --- Diana Wynne Jones,
│ _Archer's Goon_

Attachment: signature.asc
Description: OpenPGP digital signature