Kodsnack 536 - I choose computer science, with Michele Riva

Recorded at the Øredev 2022 developer conference, Fredrik chats with Michele Riva about writing a full-text search engine, maintaining 8% of all Node modules, going to one conference per week, refactoring, the value of a good algorithm, and a lot more. Michele highly recommends writing a full-text search engine. He created Lyra- later renamed Orama, and encourages writing your own in order to demystify subjects. Since the podcast was recorded, Michele has left his then employer Nearform and founded Oramasearch to focus on the search engine full time. We also discuss working for product companies versus consulting, versus open source. It’s more about differences between companies than anything else. Open source teaches you deal with more and more different people. Writing code is never just writing code. Should we worry about taking on too many dependencies? Michele is in favour of not fearing dependencies, but ensuring you understand how things important parts for your application work. Writing books is never convenient, but it can open many doors. When it comes to learning, there are areas where a whole level of tutorials are missing - where there is only really surface-level tutorial and perhaps deep papers, but nothing in between. Michele works quite a bit on bridging such gaps through his presentations. Thank you Cloudnet for sponsoring our VPS! Comments, questions or tips? We are @kodsnack, @tobiashieta, @oferlundand @bjoreman on Twitter, have a page on Facebook and can be emailed at [email protected] if you want to write longer. We read everything we receive. If you enjoy Kodsnack we would love a review in iTunes! You can also support the podcast by buying us a coffee (or two!) through Ko-fi. Links Michele Michele’s Øredev 2023 presentations Nearform TC39 - the committee which evolves Javascript as a language Matteo Collina - worked at Nearform, works with the Node technical steering committee Lyra - the full-text search engine - has been renamed Orama Lucene Solr Elasticsearch Radix tree Prefix tree Inverted index Thoughtworks McKinsey Daniel Stenberg Curl Deno Express Fastify Turbopack Turborepo from Vercel Vercel Fast queue Refactoring Michele’s refactoring talk Real-world Next.js - Michele’s book Next.js Multitenancy Create React app Nuxt Vue Sveltekit TF-IDF - “term frequency–inverse document frequency” Cosine similarity Michele’s talk on building Lyra Explaining distributed systems like I’m five Are all programming languages in English? 4th dimension Prolog Velato - programming language using MIDI files as source code Titles For foreign people, it’s Mitch That kind of maintenance A very particular company A culture around open source software Now part of the 8% Nothing more than a radix tree One simple and common API Multiple ways of doing consultancy What you’re doing is hidden You can’t expect to change people A problem we definitely created ourselves Math or magic Writing books is never convenient Good for 90% of the use cases (When I can choose,) I choose computer science