[mercer]

Don't forget find the perfect static site generator, realize you might as well use Wordpress with a nice theme, try out a plugin that does <x> badly, write <x> manually, decide to make it a plugin, learn how Wordpress does plugins (it's not great!), get annoyed by theme so build one yourself, find out what this 'css grid' thing is all about, sprinkle in some JS for something cool, use the fancy new features despite not needing them so choose a build tool that is simpler than Webpack, learn that tool and run into a weird issue so go for Webpack anyways, find out Webpack has changed completely, get everything working and decide that maybe you should just use that static site generator after all. Or maybe that new LiveView thing, but you'd have to learn a new language and framework for that. Would look good on your CV maybe? Then your RPi arrives and you start tinkering with that.

[jordanbeiber]

Oh, the CMS race. Done it a few times already. Many nights spent building custom WP plugins and trying out static site generators.

Not a single blog-post but a lot of potential knowledge.

I’ve landed in static. Either GH pages or through netlify CMS. For now.

Needs re-evaluation, I can tell. But when?! I have a family to run.

[pbhjpbhj]

I've been out of web dev for a while and am supposed to be doing a site today, check my comment history for how well that's going. My design process has resembled your comment quite closely.

[mercer]

While I do still run into this exact problem, I do think I've learned something along the way.

Basically, it boils down to: commit to either doing most of it yourself, with carefully considered bits of outsourced help, or go for an outsourced solution and deal with the bits you need to change.

The former usually is more fun, and probably works best for stuff you need to work on (as in, alter) over time.

The latter is less fun, but if you can handle working within the boundaries presented and if the thing you're doing is throwaway or relatively unchanging, it's probably the best way to go.

When it comes to CRUD web stuff, for example, I find that Wordpress is often the best way to go, even though I hate most of it with a passion. But the fact is that with just a few plugins (Advanced Custom Fields primarily!), it's often one of the best solutions available for a typical website, especially if it has news/blog type stuff and if it's not mission-critical to yourself or the client.

The same applies to smaller 'units' of code. Every time you add code that isn't your own, you add limitations and risks, but in practice it's often worth it. The skill I'm trying to improve is to know when to make what choice in this regard.

(to be clear, when I choose Wordpress, I'm basically accepting that almost the whole thing is a dumpster fire. Most of my work then involves isolating myself from the resulting mess, whether by avoiding WP's templating/querying system, or making sure a site runs on its own server. And yet it's still worth it at times)

[TeMPOraL]

> and am supposed to be doing a site today, check my comment history for how well that's going

I burst into laughter when I read this, holy shit if I'm not doing the exact same thing right now. You could probably graph my mood by HN comments frequency.

[chimen]

This is basically how I made Qards (https://qards.io). I kid you not. Good to see I'm not alone :)