Why a Table is a good idea

[felix91gr]

Because it allows us to show:

1. State of development
2. Systems supported
3. A good description: use cases, for example

All of this, without cluttering everything

So, it's something to think about. We'll try and see.

[vknabel]

@felix91gr generally I would prefer a table, too. But this is something we should test with actual data as if the table has too many columns, it may be awkward to read when mobile.

What do you think of not editing the README directly, but instead edit another file (let's say YAML) and generate the README using some Swift tooling (e.g. using Stencil and one of Beak, Marathon, Sake, Archery or a plain SPM package)?

Pros:

• the markup could be changed easily
• we could a github page with a search function in the long term
• sorting entries

Cons:

• less intuitive
• we would need to run a command in order to apply the changes (can travis do this automatically?)

[felix91gr]

What do you think of not editing the README directly, but instead edit another file (let's say YAML) and generate the README using some Swift tooling (e.g. using Stencil and one of Beak, Marathon, Sake, Archery or a plain SPM package)?

That would be really interesting, specially because the Pros you list are very real and useful. Sorting entries would be amazing. I don't know how to use any of those tools (except SPM), but I'm pretty sure you can help me with that :)

In general, I think the current setup is nice for now because it's more approachable for anyone that wants to help out (you just click "Edit" and then "Submit a Pull Request"). In the future, I think we can use a github page and figure out a way of making it as intuitive as possible (maybe a web form? dunno).

I'm gonna investigate how the deployment is, so that we can start brainstorming :)

[vknabel]

The tools I mentioned are just helpers for executing swift files or projects. I will open a PR and add those tools to the readme.

Yes, using the readme as single source of truth has huge benefits.

Instead of configs we could parse the markdown and use danger to automatically comment on PRs if the order or the markup is wrong. But this approach would require a bit more programming work.