"Michael Sullivan" <·······@bcect.com> wrote in message
····································@bcect.com...
> Seriously to Raffael, Karl's clip-it crack was on target -- we have a
> great deal of trouble writing good "conversational" software to do
> things like aid in the set up of a word-processing document. What makes
I would just caution against taking one spectacular failure as evidence that
the problem is too hard to solve. Does anyone know of any other significant
attempts at similar virtual assistants? I do know of a couple but confess
that I haven't examined them that much. One was some Purple Gorilla that
mostly liked to say hi and tell jokes and try to send you to its sponsor's
websites.
Perhaps its time is not yet here, but I do believe a successful "virtual
super-user consultant" is in the not-so distant future. This will soon
follow with some basic automated programming, scripting and database
building driven by users who do not understand the code behind the scenes.
After that...I don't know how far it can go.
It will probably follow the usual rule that the last 5% that are hard
problems will take 95% of the work. But lets face it, 95% of programming
across the industry is "code monkey" work (my apologies to simians
everywhere) And this is the 95% that will eventually be generated code. 30
years (less?) ago, who would have thought that a car could be built by a
team of robots?
--
Coby Beck
(remove #\Space "coby 101 @ bigpond . com")