The Heat Death of Prolog

[Mild Shock]

An increase in entropy indicates greater uncertainty
in the implementers's choice of operations,
potentially reflecting a decrease of intuition.

Maximum entropy occurs when all operations are
equally likely, corresponding to a state where
the implementer acts randomly due to lack of

intuitive guidance. In this framework, we might
interpret the "heat death" of Prolog as a state
where implementers no longer have effective

intuition or insight to guide their work.

AI-driven development of Prolog systems
https://lims.ac.uk/documents/undefined-10.pdf

[Mild Shock]

Hi,

The price that nobody needs:

- Alain Colmerauer Prolog Heritage Prize
recent practical accomplishments that
highlight the benefits of Prolog-inspired
computing for the future

- Theresa Swift and Carl Andersen
Janus nonsense

- Michael Leuschel and STUPS Group
ProB nonsense

https://logicprogramming.org/alain-colmerauer-prize/

The price that everybody wants:

- Alain Colmerauer Prolog Systems Price
For contributions of lasting and major
technical importance to Prolog Systems
design.

- Mats Carlsson: SICstus Prolog
https://www.ri.se/en/person/mats-carlsson

- Jan Wielemaker: SWI Prolog
https://en.wikipedia.org/wiki/Jan_Wielemaker

- Ulrich Neumerkel: ISO Standard
https://informatics.tuwien.ac.at/people/ulrich-neumerkel

- Markus Triska: CLP Integration
https://www.metalevel.at/

- Taisuke Sato: Tabulated Resolution
https://rjida.meijo-u.ac.jp/sato-www/sato/

Etc.. Etc..

Bye

Post by Mild Shock
An increase in entropy indicates greater uncertainty
in the implementers's choice of operations,
potentially reflecting a decrease of intuition.
Maximum entropy occurs when all operations are
equally likely, corresponding to a state where
the implementer acts randomly due to lack of
intuitive guidance. In this framework, we might
interpret the "heat death" of Prolog as a state
where implementers no longer have effective
intuition or insight to guide their work.
AI-driven development of Prolog systems
https://lims.ac.uk/documents/undefined-10.pdf

[Mild Shock]

So how can we decrease entropy in Prolog
systems design? Here some bad examples
that rather increase entropy:

- Janus nonsense:
Nobody cares about features terms, just
use member(x-V, [x-10,y-20]) (*), nobody needs
a copying foreign function interface.

- ProB nonsense:
Papers like "Making ProB compatible with
SWI-Prolog" rater point to problems, than
to problem solution pairs.

(*) Ok, that was sarcasm with a grain of
salt. But its folk knowledge that for for
small size dicts linear search is indeed
on par with binary search.

Post by Mild Shock
Hi,
- Alain Colmerauer Prolog Heritage Prize
  recent practical accomplishments that
  highlight the benefits of Prolog-inspired
  computing for the future
  - Theresa Swift and Carl Andersen
    Janus nonsense
  - Michael Leuschel and STUPS Group
    ProB nonsense
https://logicprogramming.org/alain-colmerauer-prize/
- Alain Colmerauer Prolog Systems Price
  For contributions of lasting and major
  technical importance to Prolog Systems
  design.
  - Mats Carlsson: SICstus Prolog
  https://www.ri.se/en/person/mats-carlsson
  - Jan Wielemaker: SWI Prolog
  https://en.wikipedia.org/wiki/Jan_Wielemaker
  - Ulrich Neumerkel: ISO Standard
  https://informatics.tuwien.ac.at/people/ulrich-neumerkel
  - Markus Triska: CLP Integration
  https://www.metalevel.at/
  - Taisuke Sato: Tabulated Resolution
  https://rjida.meijo-u.ac.jp/sato-www/sato/
  Etc.. Etc..
Bye

Post by Mild Shock
An increase in entropy indicates greater uncertainty
in the implementers's choice of operations,
potentially reflecting a decrease of intuition.
Maximum entropy occurs when all operations are
equally likely, corresponding to a state where
the implementer acts randomly due to lack of
intuitive guidance. In this framework, we might
interpret the "heat death" of Prolog as a state
where implementers no longer have effective
intuition or insight to guide their work.
AI-driven development of Prolog systems
https://lims.ac.uk/documents/undefined-10.pdf

[Mild Shock]

But lets go a back in time. I guess a
Alain Colmerauer Prolog Systems Price is
surely deserved by this implementer:

- David H. D. Warren: Warren Abstract Machine
https://bristol.academia.edu/DavidWarren

WAM is still influential, see Scryer Prolog.
He had also charming ideas about dicts:

dic(salt, sel,
dic(mustard, moutarde,
void,
dic(pepper, poivre, void, void)),
dic(vinegar, vinaigre, void, void))

http://sovietov.com/tmp/warren1980.pdf

But I guess today we would prefer dicts
that remember some input order, right?

Post by Mild Shock
So how can we decrease entropy in Prolog
systems design? Here some bad examples
  Nobody cares about features terms, just
  use member(x-V, [x-10,y-20]) (*), nobody needs
  a copying foreign function interface.
  Papers like "Making ProB compatible with
  SWI-Prolog" rater point to problems, than
  to problem solution pairs.
(*) Ok, that was sarcasm with a grain of
salt. But its folk knowledge that for for
small size dicts linear search is indeed
on par with binary search.

Post by Mild Shock
Hi,
- Alain Colmerauer Prolog Heritage Prize
   recent practical accomplishments that
   highlight the benefits of Prolog-inspired
   computing for the future
   - Theresa Swift and Carl Andersen
     Janus nonsense
   - Michael Leuschel and STUPS Group
     ProB nonsense
https://logicprogramming.org/alain-colmerauer-prize/
- Alain Colmerauer Prolog Systems Price
   For contributions of lasting and major
   technical importance to Prolog Systems
   design.
   - Mats Carlsson: SICstus Prolog
   https://www.ri.se/en/person/mats-carlsson
   - Jan Wielemaker: SWI Prolog
   https://en.wikipedia.org/wiki/Jan_Wielemaker
   - Ulrich Neumerkel: ISO Standard
   https://informatics.tuwien.ac.at/people/ulrich-neumerkel
   - Markus Triska: CLP Integration
   https://www.metalevel.at/
   - Taisuke Sato: Tabulated Resolution
   https://rjida.meijo-u.ac.jp/sato-www/sato/
   Etc.. Etc..
Bye

Post by Mild Shock
An increase in entropy indicates greater uncertainty
in the implementers's choice of operations,
potentially reflecting a decrease of intuition.
Maximum entropy occurs when all operations are
equally likely, corresponding to a state where
the implementer acts randomly due to lack of
intuitive guidance. In this framework, we might
interpret the "heat death" of Prolog as a state
where implementers no longer have effective
intuition or insight to guide their work.
AI-driven development of Prolog systems
https://lims.ac.uk/documents/undefined-10.pdf