Log #14: I don't like draft one
After looking at the first draft of the info product in styled PDF form, I’ve decided I don’t like it. Before getting into that, let’s look at how I’m putting together the PDF.
Last time I mentioned trying asciidoctor.
I gave it a quick try and it seems like a nice tool, but I don’t really want
to learn a new type of markup (and convert from more vanilla markdown) right
now, so I decided against it and looked into other options.
There are a few different tools that will generate a PDF from markdown.
You can use pandoc, which I believe uses some flavour of LaTeX under
the hood.
(I’ve thought about just using LaTeX outright, but I fall into the same trap
of moving from markdown.)
But I want more control over the styling and I’m only really interested in
PDF files for the first release.
After some more research, I found quarto, which seemed pretty good - I like
how you set up the document.
And then I found out that it uses pandoc and LaTeX, so the only benefit
over doing it yourself with pandoc is the build system and a few other bits.
Styling is still opinionated, so not quite what I want.
Basically, I just want a simple way to build a PDF from some source markdown files and style it.
The last option I could think of is to build the PDF by turning the markdown into a static site (with Jekyll, since that’s what I’m familiar with) and then export it as a PDF. It’s a weird way of handling it, but it’s basically how I write now, it has familiar styling and – if you use Playwright to capture the PDF from the HTML page – you don’t have to worry about outdated CSS engines some PDF tools use.
I put together a quick proof-of-concept and it works, though it’s not perfect. I’m sure the bespoke method of handling it will come back to bite me at some point, especially if I decide to publish in other formats. But I’m happy with it for now.
The goal of the PDF stuff was to take the content for draft one and put it together to take a look at it in context.
I didn’t realise I’d written nearly 100 pages, so that was cool to see. And it looks nice enough, so we only need a few tweaks there. But I’m not happy with the first draft. I don’t think I’m ever happy with the first draft of anything. What are the issues? I’m not 100% sure yet. I need some time to let it digest before coming up with steps for draft two.
In the meantime, I’ve neglecting other stuff, and that isn’t sitting well with me either. So my current plan for May is to write and publish and article while lowering the priority of the info product with the aim to ramp back up and publish in June. That’s still an ambitious goal…
I read a tweet or a comment or something earlier this week that was along the lines of ‘If I see AI-generated images as part of content anywhere, I dismiss it as low effort and low quality’, and I think that’s where I currently stand on the issue too. If I can tell that an AI generated anything, text, images, video, I immediately lose interest and skip it – it’s almost certainly a waste of time. I’ve always worked on the notion of ‘publish only what you would consume personally’, so I need to find a good artist to help with the images. This is where I wish I had a larger network or a large following online where I could just tweet “does anyone know a freelance artist that can do something in this style?”
The desk chair I ordered over six weeks ago is still stuck in purgatory. After hounding them for more details, it turns out that the issue is with a logistics company that hasn’t been able to arrange delivery because they have no availability to deliver in my area, or something like that. It feels like a weak excuse. I’ve been given a small refund for the inconvenience. Will I receive it before the two month mark?
What’s next
- Think about path towards draft two of the info product.
- Start writing May’s article.