SBCL20 in Vienna
lispblog.xach.com
lispblog.xach.com
Notably, the venue of this conference was the Bundesrechenzentrum (Austrian Federal Computing Center), abbreviated as BRZ, the strategic technology partner of the Austrian public sector where many IT applications and e-government services are developed and hosted:
I work in the public sector and have had the great pleasure to work with Philipp Marek, who hosted this conference, and many of his colleagues already in several projects where Lisp was used to achieve impressive and also fast results.
Currently, I am leading a project in the Federal Ministry for Digital and Economic Affairs where — in close cooperation with the BRZ as our contractor, and also with other ministries and organizations — we are using symbolic AI techniques to deduce and show suitable grants for businesses, based on logical rules and criteria. The eventual goal of these developments is to automatically suggest fitting grants in the Business Service Portal (usp.gv.at), the Austrian e-government platform for businesses.
The BRZ team presented a first draft of the rule-based DSL for encoding eligibility requirements, specified as Lisp forms, already in the project kick-off (i.e., first) meeting. This shows that Lisp is very suitable for fast prototyping, and will be an excellent language to implement the logic we need in this project.
In the public sector, we are especially interested in symbolic AI methods, to ensure traceability and explainability of decisions. Lisp and Prolog programmers are especially needed for these projects, and it is great to see that the BRZ fosters these developments by hosting such conferences and already using these languages in several projects.
However, Lisp has already played an important role in two projects I led, and it also plays an important role in the project I mentioned which is currently ongoing.
Prolog is used for example in a theorem prover for deontic logic developed by Björn Lellmann at TU Wien:
http://subsell.logic.at/bprover/deonticProver/
In cooperation with Björn, we have used this solver to showcase the formalisation and automated application of legislation. For a prototype, Björn has formulated a few paragraphs of the Studienförderungsgesetz (a law regulating student grants) in deontic logic, and demonstrated that this theorem prover can be used to automate various decisions because they can be deduced as stringent logical consequences of existing regulations.
Generally, I think that as the tasks we want to solve become more challenging and encompassing, high-level languages such as Lisp and Prolog will be increasingly needed.
If any Lisp hackers feel like moving to Porto or Lisbon, give it a shot.