On Tuesday 13 March 2007 07:26, Eric Baum wrote:
> Is there some reason why it is not the most natural thing
> to look at the Helevetica Reader (as with pretty much any proper
> noun) as an instance in the
> class of font readers? It inherits pretty much everything from
> existing font readers, except a new method or methods (which
> themselves are refinements of old methods) for recognizing
> text.
"Object-oriented programming is an exceptionally bad idea which could only
have originated in California."
- Edsger Dijkstra
To me, the OOP metaphor way over-complicates things. When you learn a new
font, your new module feeds off of the same lower-level edge-finders and
whatnot, and feeds into the same higher-level word and phrase recognizers. It
passes the same reverse-flow expectations traffic (from above: what word
might be next? to below: look in that spot just to the right of *here*).
In an evolutionary account of where these modules came from (or a market one),
there is clearly inheritance. But you seem to be implying that there would be
a Reader object that describes the whole shebang and specifies the position
and connections of the new character-recognizer in it. I think the decision
to split/copy the character recognizer is almost certainly local and would
not involve a formal higher architecture spec. It would get whatever feedback
it needed in the form of something akin to price signals, allowing a global
Pareto optimality from purely local decisions.
BTW, I can see how a system that could manage/redesign the whole shebang would
be USEFUL, but I just don't see how it could work short of having already
built a working AI that is a competent systems analyst.
Josh
-----
This list is sponsored by AGIRI: http://www.agiri.org/email
To unsubscribe or change your options, please go to:
http://v2.listbox.com/member/?list_id=303
