Data and Markup
A few years ago I did the LDL project. Mostly unused, but an interesting experiment. Lately I've been giving a lot of thought to data and markup.
The optimal structure of the two seems opposed. Data is best organized hierarchically with each node being either a container (primarily dictionary or list) or data. Markup also follows a tree structure, but data - instead of being a specific node - gets wedged in between the structure.
JSON seems to be a pretty ideal standard for data. I like the Media wiki for mark-up, but it's not very extensible, it's very context sensitive and uses up a lot of characters.
Question: How could Data and Markup be combined either into a single standard to two halves of a standard that play well together.
Also consider
- transforms
- tables
- print face
- human / computer readable
- image/blob
11/30/15 Notes
The important bit is to have something that can translate to all the other formats. That's always the winning feature.
- html
- MediaWiki
- JSON
- Github Markdown
- Recursive: From Go, I like that there is a standard way to do things, having the processor handle the standards would be nice. Also, good for cleaning up tables.
As stated above, the structure will be hybrid of data and markup. I think I want to add a CSS like sub-language for defining custom tags.
- Headings: ==, but has to start the line.
- Lists: I'll need to get used to it, but I'll do -, also, if every line in a list has : in it, and the portion preceding : is unique it becomes a keyed list
- Links: we'll do square brackets, if there's a space, everything after is the link text, otherwise, it's the link.
- Internal anchors: how to set and how to link
- ID & Class: Probably like I do everywhere else #id.class1.class2, needs to start a line, applied to the next item
- Tables
- Image
- Tagging