Historia de las APIs

Siempre existen conversaciones de APIs, desde el boom de los microservicios se le dice coloquialmente APIs donde se refiere por default a API Rest o Web API, sin duda, los mas senior solo piensan en la API de C, C++, Java, etc.

Dicho esto es simpatico cuando aparece por primera vez ese termino.

Todo apunta que fue en:

• AFIPS Spring Joint Computer Conference: Data Structures and Techniques for Remote Computer Graphics, I. W. Cotton y F. S. Greatorex Jr., 1968

Finally, hardware independence at the central computer means that a consistent application program interface could be maintained if that computer were replaced...

Ademas, los siguientes libros y presentacion hablan mas de la historia de las APIs:

• A Brief, Opinionated History of the API, Joshua Bloch (InfoQ, 2018) 
• Building an API Product Design, implement, and release API, Bruno Pedro, 2024
• Enterprise API Management, Luis Weir, 2019
• Learning API Styles_ Understanding the Trade-Offs of Common, Lukasz Dynowski & Marcin Dulak, 2025

TL;DR

Sigue usando API para cosas modernas estilo API Rest, y ponle el adjetivo si quieres hablar mas claro de otro tipo de APIs, como:

EDSAC subroutine library, FORTRAN libraries, IBM System/360 interfaces, Application Program Interface (1968), Unix system calls, C Standard Library, POSIX API, MS-DOS API, Macintosh Toolbox, Windows API, Xlib, RPC, OpenGL, CORBA, Java API, SOAP Web Services, .NET API, REST APIs, GraphQL, gRPC.
 

Algorithm Recreational Programming

Newsletter de inicio de los 90s que es como la continuacion del articulo de Scientific American que desde matematicas recreativas paso a programacion recreativa.

Amdahl’s Law

Amdahl’s Law can be expressed mathematically as follows:

$$ \mathrm{Speedup}_{\max} = \frac{1}{(1-p) + \frac{p}{s}} $$

• SpeedupMAX = maximum performance gain
• s = performance gain factor of p after implementing the enhancements.
• p = the part whose performance needs to be improved.

Let’s take an example. If the part that can be improved is 30% of the overall system and its performance can be doubled, then:

$$ \displaystyle \mathrm{Speedup}_{\max} = \frac{1}{(1-0.30)+\frac{0.30}{2}} = 1.18 $$

Oregon State University — Amdahl’s Law

If you buy a system with n cores, you should get n times Speedup (and 100% Speedup Efficiency), right? Wrong! There is always some fraction of the total operation that is inherently sequential and cannot be parallelized no matter what you do. This includes reading data, setting up calculations, control logic, storing results, etc.

Tutorialspoint — What is Amdahl’s Law?

Wikipedia — Amdahl’s Law

La Ley de Amdahl dice que el rendimiento total está limitado por la parte del sistema que no podemos acelerar.