Re: Fitting PR and RR into logical rules

Dear Sandro,
>> Once again, I would like to express a very strong warning against
>> further considering interchange between production/eca rules and
>> deduction rules.
>>
>> What I read oin this here is, I am sorry to say it openly, is too
>> superficial.
>>     
>
> Superficial is good.  Superficial allows widespread adoption.  Unix was
> superficial compared to competing operating systems.  HTML was insanely
> superficial compared other document formats.  HTTP was terribly
> superficial compared to other data distribution protocols.
>
> If RIF is superfical but still just good enough to addresses 60% of the
> rule use cases (and uses its superficiality well -- ie as simplicity) it
> stands a good chance of global success.
>   

I am afraid you misunderstood my comment. I hesitated before choosing
the word "superficial". Let me be more precise:

Thinking of an interchange between deductive and production/eca rules is
in my humbleopinion naive because it raises extremely complicated issues
that are far from being solved and most probably never will be.

Otherwise, one af the great dreams of software engineering would be
achieved: automatically generating a imperative program from a
declarative specification or vice versa.

Unix nevewr has been superficially conceived.

HTML in its first version was a desaster -- that would have been avoided
by looking a bit at the available literature. It took about a decade to
bring HTML to an acceptable state.

HTTP was simple. This is right.

I was not speaking of the simplicity of the proposed solution but
binstead of the danger for us, designers of RIF, to be too short
thinking or naive. Right? :-)

(I love extremely simple solutions and the paper for which I am a bit
known describes one such solution.)

Francois
>      -- Sandro
>

Received on Friday, 2 June 2006 13:01:01 UTC