Review: The Mote in God's Eye 👍

I just finished The Mote in God's Eye by Larry Niven & Jerry Pournelle.  It was excellent.  It's my favorite kind of speculative fiction: imagine one small change to how the world works and then explore its consequences.  In this case, the exploration is packaged as a first contact narrative which makes the comparative anthropology feel natural and entertaining.

Ignoring superficial differences, alien biology deviates from human biology in one way.  That deviation leads to substantial differences in their evolution, societal organization, priorities, and motives.  Niven and Pournelle toss in a few other interesting ideas along the way.

I sometimes had trouble keeping the cast of characters straight.  I've never had a mind for large social networks, so that's probably as much my problem as theirs.  Either way, it didn't detract from the story or the ideas.  It's definitely worth reading.

Databases by code size

I was curious how various databases compare in terms of their total source lines of code.  For each database I cloned the source repository or downloaded a source tarball.  I deleted code related to tests, and producing documentation (some projects have more test code than database code, so this can make a big difference).  Then I ran scc.  The numbers below are what scc labels as Code.  Since the methodology isn't exact, I only retained two significant figures.

The results are sorted from smallest to largest by SLOC count.  Let me know if you find mistakes or want another database included in the results.

• LMDB commit 332718f1 - 10k
• LevelDB 1.22 - 13k
• Tokyo Cabinet 1.4.48 - 40k
• GDBM 1.18.1 - 41k
• Kyoto Cabinet 1.2.77 - 61k 
• QDBM 1.8.78 - 81k
• Redis 5.0.5 - 140k
• Berkeley DB 5.3.21 - 180k
• RocksDB 6.2.2 - 190k
• SQLite 3.29.0 - 318k
• Neo4j 3.5.8 - 430k
• Cockroach DB 19.1.3 - 630k
• PostgreSQL 11.5 - 860k
• MariaDB 10.4.7 - 2,200k
• MySQL 8.0.17 - 3,700k
• MongoDB 4.0.12 - 4,100k