Post

As most of my readers will probably know, at my day job, I help develop Glamorous Toolkit . And, I’m proud to say, after two years of working on and within it, it has thoroughly infected my computing life. I rarely work in the terminal anymore. I don’t even miss it. In fact, when I go back, it feels a bit like I am transported back into a pre-historic age.

But I didn’t write the first blog post in almost a year to sing the praises of a tool that many of you probably haven’t tried and a reasonable portion probably will never try. I’ve come to talk about how I now write my blog posts from within Glamorous Toolkit.

Glamorous Toolkit comes with a knowledge management system called Lepiter. It is not unlike Roam Research or even Jupyter notebooks, except that it is moldable . That means that users can extend it to house their own types of content and snippets. Like in a Jupyter notebook you can interleave code and text. But unlike a notebook, I can easily extend it to house arbitrary snippet types, for instance YouTube video embeds. It is a fun environment to write, think, and work in. Most of our new code gets its start as a Lepiter document, where we play in snippets and add views until the code base has formed, almost as if by accident. It is hard to describe, but I was thinking to do a live-coding session or two when I find the time.

Anyway, Lepiter gives me the first step for a publishing tool. I have the document. Now I just need to export it.

Luckily we’ve lately been working on a solid HTML exporter that is now good enough to export all kinds of documents. With it, I might be able to publish this blog post. Let’s see how we could do it.

From here on out, this blog post will be a literal document. In fact, it’s starting out that way as I write: I’m writing my blog post, and at the same time I’m writing the code to export that blog post. First, I need to get to the page.