As always: test on the data.

XZ stays my favorite though.

Why benchmarks should be carefully used as arguments

The benchmarks compare compilation time, runtime of major softwares, encoding and decoding of media. I see a 10x improvements between 2008 and 2026.

Redis is faster, but they have the same order of magnitude. Maybe +1 order for Postgres in same examples.

The comparison of a lot of data does not made though.

A bunch of Raspberry Pi are not optimal to run an AI cluster. A Framework computer crunch the numbers compared to the PIs

How fast can we recognize a word from a small pre-determined set?

This is limited on the best runs though. Interesting to keep in miind.

Interop 2022 is a benchmark, agreed on by representatives of three major browser implementations, and developed through a process of public nomination and review with input from supporters Apple, Bocoup, Google, Igalia, Microsoft, and Mozilla.

It focuses on 15 areas identified by developers as being particularly troublesome.

The goal of these multi-year interoperability efforts, in the form of Compat 2021, Interop 2022 and much more, is to fully acknowledge and address the pain points developers have experienced through for many years. And it's not a one-browser effort but rather a strong collaboration between all major browser vendors and friends for improving the web platform across the board.

Small but efficient post !
Rust is between C and C++, close to the c++.

Maybe programs contributed to the project as part of a concerted language advocacy and marketing strategy, benefit from more programmer hours spent on program optimization.

We don't know yet.

for version 7.2.4