Re: Bob's 'Self-aggrandizing ignorant' Count: Was: What databases have taught me

Dmitry A. Kazakov wrote:

> Which does not make it unpredictable, if specifically 'non-deterministic'
> was meant. If something like 'halting' was, then sorry, but all languages
> are in the same position [was it about a language?] If that was about
> non-functional constraints (like real-time issues), then, well, RA is not
> in a beacon position here.
>

I take non-determinism to mean that given an identical environment and identical input, multiple outputs are possible and are equally valid. So, no, non-determinism is not what I meant. (That said, I'm assuming that the individual objects are designed and built to be deterministic.) I meant unpredictable, in that the type of inputs and their sequence and timing can't be predicted ahead of time, therefore the behaviour of the larger state machine is unpredictable. No magic there.

All Turing complete languages are prey to the halting problem.

> DBs are stateless, I presume...
>

Did I mention DBs ? (Apart from a well-deserved swipe-in-passing at OODBMSs ?)

> I wouldn't expect such statements from people claiming themselves champions
> of theoretical formalism.
>

I don't think I'd be brave enough to claim myself a champion of theoretical formalism - my chops simply aren't up to that job. I will say however that I do not want to use in anger any tools where *nobody* has been a champion of theoretical formalism in their design, development and/or growth, and so far as I'm concerned at the moment, OO falls flat into that category. Received on Sat Jul 01 2006 - 01:35:15 CEST