I recently saw a video touting Zed (don’t you dare) as the new hotness IDE. The creator was someone who I respect a ton, but I was quite taken aback by some of the features they were demonstrating as though they were a big reveal. It went something like: “Ta-da! You can use vertical panes!” or “Ta-da! A fuzzy finder searches through your project!”.
Looking through the comments, I noticed a lot of positive engagement that made me wonder how something that I thought was relatively “solved” in Neovim could evoke such reverence and awe with a newer IDE.
Blogs
Intro# Modeling data can be the most challenging part of creating a new app. In fact, it’s been suggested that it’s totally acceptable to -not- fail fast when scoping your schema. Getting your data modeled correctly from your Initial Commit can save you hours of refactoring.
I was recently prototyping a new greenfield application for work, and as I mapped out my models and how they’d be associated, I started to experience some friction.
Another great book by David Bryant Copeland.
Where I had trouble in the past# I purchased the PragProg book Docker for Rails Developers several years ago. It was good, but a little dated considering recent docker practices, and I shelved it after running into some speedbumps (it used some webpack/er/ tooling that was hard for me to get running).
Not too long after that, I ran into some issues with a friend’s legacy app.
I recently picked up a short technical book by David Bryant Copeland to help understand Sidekiq a little better. TL;DR - The dev environment made it the most efficient technical book I’ve ever read.
Dev Setup# Technical books will never be the same for me. Initially, I ran into a few issues with the docker compose setup. After very luckily finding the forum post where Copeland himself responds and updates, I was able to get the environment setup.
Most of the consensus around AI pair programming lately has been that it’s not worth your time. Bertrand Meyer put it pretty bluntly regarding ChatGPT. Perhaps in the narrow scope of his use-case it didn’t help him, but I’ve already seen plenty of evidence to the contrary.
What it’s Not (Yet?)# Copilot and ChatGPT haven’t yet formulated a business idea, created an MVP, deployed it, and garnered seed revenue for scaling.
Just a quick blog post to spread the good word. Consider -
The unexamined editor is not worth integrating
I’m a big proponent of neovim, and I recently got a few plugins configured, and I don’t think there is any way I could work without them now.
Copilot ** AI Pair & Autocomplete Bot Mason ** LSP Manager Luasnip ** Custom Snippets Nvim-Cmp ** Autocomplete Menu All setup can be found in my dotfiles repo
Following the acquihire late last year by Shopify and some strong praise for the Remix framework via Twitter, I clocked it as a high priority for my next learning endeavour. I started my software journey with Ruby and Rails so I have a penchant for the opinionated, and fittingly I felt like Remix deserved a bit of my attention. I spent a bit of time taking Remix for a spin, and decided to consolidate some of my thoughts.
Having spent nearly a decade in IT professionally, I am a bit ashamed to admit that I never felt like there was a good time for me to “break” into the Software discipline. I was always worried that as soon as I began to learn JavaScript or Python or Bash, the next big thing would become popular, and I would have completely wasted my time learning x, y, or z. From outside of Software it felt like new magical technologies were popping up everyday (they kind of are), and it was just a matter of time before something huge completely reshaped how we access our bank accounts, schooling portals, and social media.
Although I’m sure much of the shift in apprehension towards tracking and data collection can be attributed to the “Social Dilemma” effect, it’s great to see more people thinking critically about what many “free” social media apps and services are extracting from their userbase.
It’s certainly been done before, but Apple’s big efforts to offer consumers a little more control of their data couldn’t have come at a better time.
+Taking the Plunge - NVim# The accessibility of VSCode is pretty unbelievable. There is a reason why 71% of developers surveyed on Stack Overflow use it. It has leagues of community support, and looks like it’ll be the predominant IDE for the foreseeable future.
But recently I decided that I needed something different.
When is it time to upgrade your toolset? I came to the realization that the limiting factor on how quickly and efficiently I could edit was my knowledge of VSCode.