I’ve recently read a post about how GitHub writes blog posts. They are running some tests (!) on the posts. Fascinating.

I’ve immediately decided to implement it here, on my own blog. The author provided a gist of what they are using. I’m not a Ruby guy, and I’m afraid I’m missing some fundamental knowledge about gems and bundler and stuff. But I did my best. Didn’t work. So I had a closer look at what it is doing, to see if I can implement its features in another way quickly, or if I can live without them.

The things it does

Image alt tags

OK, it’s nice, push it down in the backlog.

Image size

I don’t really care right now, maybe later.

Image promotions

Come on! This blog will barely contain 5 images anyway. I don’t need it.

CDN hosted images

The whole site is hosted on a CDN. I don’t need it.

No emoji

I agree, but it wouldn’t come to my mind to insert one either. I don’t need it.

No “Today, “

Well, it doesn’t seem that important, does it? I don’t need it.

So I could strike through the whole list and started to think of…

What do I need exactly?

Spell check.

English is my second language, and I’m not nearly as good at speaking or writing it as I want to be. Spell check would be nice.

Google told me there is a command line utility called Aspell. It is very nice, by the way, I was impressed. It can do it’s job in multiple languages, can use custom dictionaries, and shipped with my distro by default.

So I wrote a little shell script and put a .travis.yml together, and voilà, I have Continuous Integration for my blog posts.

How cool is that? I should have thought about it by myself.

I’m planning to add

  • grammar check
  • broken links check
  • image alt tags and size checks after I’ve added at least one image

So there are a number of things that has yet to be done, but I’m fine with this setup for now.

Thank you, GitHub and Zach Holman.