Mike Levin SEO

Future-proof your technology-skills with Linux, Python, vim & git... and me!

← Why Linux Fights Obsolescence and LXD is Key
Why Pages Aren't Usable On Mobile, Fixing! →

Choosing Best CSS Framework For Github Pages and Jekyll

by Mike Levin

Monday, August 15, 2022

Let’s push and push and push on that YouTube front. Wrap in Reddit. Twitter, Facebook and all that is nothing compared to the YouTube/Reddit front. I’ll maybe announce my videos on Twitter and keep the old Levinux crowd in the loop with the MikeLevinux page, but that’s out of consistency, loyalty to those audiences and helping search a wee little bit.

Oh, search. Yes, I’m always reaching out to people who search and not surfers. It’s seekers I’m seeking and not slaves to the YouTube algorithm. The YouTube algorithm has rarely been kind to me and I should assume it never will. It’s for those who predict and pander. Ha ha, and that’s from an SEO who makes a career out of showing others how to do exactly that. Perhaps I should practice more of what I preach just to bank a little more online success and keep my skills sharp. Hmm, yes maybe I will. Why am I pushing on the YouTube front if not for that? Oh yeah, it’s more for attaining levels on the Maslow hierarchy of needs. I think I’m seeking “my people”.

Here’s a journal entry that started out on the private side and I moved over to the public side because this is precisely the sort of process-sharing I wish to reside here. There’s plenty of weakest link in the chain analysis to do for my YouTube, Web and social media presence. The main thing is that I’m working on my tooling an thought-processing and general skills. I’m NOT working on actually boosting my public-facing presence. The two probably should go together more than I’ve liked to believe recently, and so to have credibility on the tooling, thought-processing, skills front I need more credibility on the public-facing front.

Sighhh, okay so be it.

1, 2, 3… 1?

I need even better organization on the public-front that performs better in Google Search and reflects the current actual state of my thinking. It needs more flexibility and less fringe design wackiness. And if I’m going mainstream, I should go as mainstream-but-timeless as I can. That means not Google’s Material UI nor any of the cool-kids modern JavaScript frameworks. Instead, I should probably stick to those that will give me a modern look, be most search-friendly, been around the longest, and perhaps allow me to produce the most clean, vanilla HTML code. Specifically bootstrap.js, the old Twitter original library that kicked off the movement, comes to mind. Does it meet the criteria? Will it force me to use sass pre rendering tools? If it does, will Jekyll do the heavy lifting for me (it’s pre-rendering anyway).

Quick googling shows Bootstrap vs. Foundation, Bulma, Semantic and UIkit. Look at which uses not class propagating HTML and is friendly with Github Pages and Jekyll. I want as simple, flexible and future-proof as possible. Ugh, the tutorials on the topic are ugly. All this stuff is ugly. I’m not thrilled with how the Web works with HTML, the DOM, JavaScript and CSS. But that’s the downfall of every tool, evolved, reality-based and compromised. Idiomatic approaches are only clear in hindsight and by that time the pee is in the pool and isn’t coming out. Okay, so live with it. Which is the solution I can live with most happily over the years? Chosen tools become a part of you, so choose carefully.

The best approach is to throw a throwaway domain to the cause. Put something out with the least code. Use elements to do the styling. Anything inside a nav element gets styled as the nav I define. I shouldn’t have to liter everything up with btn classes, yuck! Less code. More abstraction. Less classes, more purposeful elements.

Foundation! Ugh, it only took me minutes looking at Bootstrap and all those btn-this and btn-that’s to become fatigued before I even downloaded the includes. Do not fall into the btn-trap! The word is semantic markup. Stick with the most clean, semantic, plain-vanilla HTML as you can. That’s future-proofing. Boostrap is the opposite of future-proofing, mixing all its special implementation code in with your main content.

Beware rabbit holes! If any of these CSS frameworks lead you down a rabbit hole, back off and rethink!

Okay, follow the love. I’ve got a few competition projects right now, including ones I have to get done for work. Is there a really quick one-off I can do? I need an experimental sight where I can experiment with CSS frameworks under Jekyll without impacting my main site. I should also point out why this is one my mind.

Let’s develop the lpvg.org site. It’s one I need to develop soon/fast to lead this movement anyway. Do the first baby-step of this project then back off and do different work. 1, 2, 3… 1? Go visit site and see if there’s anything there. No, not even my CMS system. Perfect, so it’s a fresh start. 1, 2, 3… 1? See if I have a repo yet. The folder is there, but it’s not a repo yet. Make it a “Hello World” Github Page.

Think about organizing your recently deployed categories feature of your blogging software. Life is a lot of little journeys. Here are some of mine you might be interested in. That’s not for this project, but that’s definitely something to keep in mind. Drop it into your todo list at the bottom of this document.

Ugh, okay, do the baby steps. I activated lpvg.org on Github pages. I put a _layouts folder in place and a default.html file in location. I stripped it down to the bare essentials.