Prolog at 21 on TIOBE index

Nice to know
1

TIOBE index. (March 2022 - Prolog - 21)

Looking back can you believe that Prolog was 3 on the list in 1987.

image
image936Ă—663 6.17 KB

3 Likes
2

In 1987, I was working on IBM Prolog. :wink:
#1 was C, which was fairly new then (I was TL for IBM’s mainframe C library); #2 was Lisp (and special-purpose hardware was popular - SPARC even had some of its design targeted towards Lisp); #4 was C++ (wasn’t it still using cfront then?); #5 was Basic.

And then came the “AI winter” … are we seeing an “AI spring”?

Interesting that today Prolog ranks above COBOL and Fortran – there’s a lot of legacy software in COBOL (banks, etc.) and a lot of statistics stuff in Fortran (although often wrapped in R).

3 Likes
3

Prolog broke into the top 20 again a few months ago (should have captured it then for historical use). As of June 2022 it is still at 20.

image
image927Ă—772 28.7 KB

5

If Prolog were in the top 5, then we would be in a machine-led utopia by now :grinning:

1 Like
6

“Ruby on Rails” is a terrible design (doesn’t support “undo”, journalling, etc. AFAICT).

But a persistent database of tuples might be a winner … I’m doing a bit of work on @jan’s experimental predicate store using RocksDB; it should be possible to expand this to work with other databases. But first, I want to modify some of my code to use the persistent database and see if it’s performant enough.

8

Yes, that’s what I don’t like about Rails. My experience with production databases is that directly manipulating the database is often a bad idea (and not just because it doesn’t provide “undo”); and object-relational mapping has its own set of problems (again, from bitter experience).

Rails provides a fast way to get things up and running with a persistent data store; but, if anything, it gets in the way of good database design. (Some day, I should write a blog post about good production database design and why programmers should all learn best practices from how double entry bookkeeping is done …)

Anyway, despite my misgivings, something like a Rails integration with a persistent datastore might be a “killer app” for Prolog, especially as Prolog is a far better query language than SQL.