> ... what criteria should apply [to complexity] and why one ranks
higher than another.

Don't have a chance to look up the source, but Dijkstra has a
wonderful observation about every implementation of a solution in CS
includes two kinds of complexity: that intrinsic to the problem, and
that introduced by the tools you're using to express it in a way
computers can interpret.

While not a well defined measure, I think this is a powerful intuition.

- Andrey

Sent from my cell. Please forgive abbreviations and typos.

On Feb 26, 2010, at 7:50 PM, Gerry J <[email protected]> wrote:

> John, et al
> I am interested in what you think are the better approach
> alternatives to handle complexity and size (etc), what criteria
> should apply and why one ranks higher than another.
> For example, should a language support both actors and be model
> driven?
> Is a mix of type inference and explicit typing with operators (like
> OCAML) better than extremely late binding, and for what?
> Should there be a hierarchy of syntax compatible languages, with
> different restrictions, say extremely late binding at the top, and
> fully typed and OS or device driver oriented at the bottom?
> (ie pick the right tool in the family, size of hand held screwdriver
> up to exchangeable bits for a power tool).
> Thanks for your interesting references and insights.
>
> Regards,
> Gerry Jensen
>
>
>
>
>
> _______________________________________________
> fonc mailing list
> [email protected]
> http://vpri.org/mailman/listinfo/fonc

 > ... what criteria should apply [to complexity] and why one ranks
higher than another.

Don't have a chance to look up the source, but Dijkstra has a
wonderful observation about every implementation of a solution in CS
includes two kinds of complexity: that intrinsic to the problem, and
that introduced by the tools/lan

- Andrey

Sent from my cell. Please forgive abbreviations and typos.

On Feb 26, 2010, at 7:50 PM, Gerry J <[email protected]> wrote:

> John, et al
> I am interested in what you think are the better approach
> alternatives to handle complexity and size (etc), what criteria
> should apply and why one ranks higher than another.
> For example, should a language support both actors and be model
> driven?
> Is a mix of type inference and explicit typing with operators (like
> OCAML) better than extremely late binding, and for what?
> Should there be a hierarchy of syntax compatible languages, with
> different restrictions, say extremely late binding at the top, and
> fully typed and OS or device driver oriented at the bottom?
> (ie pick the right tool in the family, size of hand held screwdriver
> up to exchangeable bits for a power tool).
> Thanks for your interesting references and insights.
>
> Regards,
> Gerry Jensen
>
>
>
>
>
> _______________________________________________
> fonc mailing list
> [email protected]
> http://vpri.org/mailman/listinfo/fonc

_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc

Reply via email to