Daniel Lemire's blog – Daniel Lemire is a computer science profes...

Imagine a world where becoming a doctor isn’t about years of rigorous study, but about showcasing your life’s work. That’s how doctorates used to roll. You’d write a book, make a groundbreaking discovery, and voila, a doctorate was yours. Fast forward...

C++ programmers tend to represent strings using the std::string class. Though the implementation might vary, each instance of an std::string might use 32 bytes. Though it is not a large amount of memory, it can add up.
In the Node.js runtime, as part of the build tools, there is a function which precomputes the string representation of all 16-bit integers, followed by a comma....

We sometimes have to work a large quantity of floating-point numbers. This volume can be detrimental to performance. Thus we often want to compress these numbers. Large-language models routinely do so.
A sensible approach is to convert them to brain floating point numbers. These are 16-bit numbers that are often capable of representing accurately a wide range of numbers. Compared to the common 64-bit floating-point numbers, it is a net saving of 4...

Most programming languages support floating-point numbers. You typically have the ability to turn a string into a floating-point number. E.g., “3.1416” could be parsed as a number close to pi. However strings typically cannot be represented exactly or...