Create a Rosetta Stone for wardley maps

[juliusgb]

Rosetta Stone: a set of examples showing how to write the same map in multiple different mapping tools.
E.g., here's map-a. This is what it looks like this in dsl-1, dsl-2, dsl-3. And that's our Rosetta stone.

Sources of inspiration:

• https://text-to-diagram.com/
• https://docs.kroki.io/kroki/

[tasel]

Ideas:

• Use an enhanced version of the tea shop example
• Create the map in Miro as its not limited
• redo the map in various tools / DSL

[padajo]

I think the interesting thing here is that I think we focus too much on what the final map "looks" like. Simon has often said things like the axes are scaffolding and unnecessary, yet most people use at least the x-axis.

I've tried to distil what I understand as a map down to things like a needs b and b needs c and b needs d and really the only thing missing from that is that a is the anchor and the relative evolution of b, c and d.

I think we overcomplicate it.

Hence why I wrote this blog post.

The elements like capital flows and PST are then layers on top of this. That is something I haven't quite got to in terms of how to write in a DSL but it's not overly complicated.

[padajo]

4 simple rules for simple Wardley maps

On a Wardley Map:

1. vertical (see [2]) and horizontal (see [4]) position have meaning, but it is only relative to other elements of the Map
2. “needs” relationship are modelled vertically upwards
3. “anchors” are “components” that have no needs
4. each “component” on a Wardley Map needs to have an evolution “score” relative to all other “components” on the same map and this is modelled horizontally increasing (“more evolved”) to the right

[juliusgb]

@padajo The background is that we were working on what to standardise and how. After a few discussions, these 3 emerged (see image below)

Since syntax (marked red in image) for representing wardley maps hasn't evolved that much, attempting to standardise it right now would also slow down experimenting with other syntaxes.

That led us to the Rosetta Stone idea. As many DSLs come up, we'll eventually converge to some kind of convention. That leaves us time to focus on the other 2 areas.

Your example can be its own DSL that we add to this Rosetta Stone

a needs b and b needs c and b needs d and really the only thing missing from that is that a is the anchor and the relative evolution of b, c and d

[padajo]

So capturing a range of DSLs is the initial goal here rather than anything.

I like that.

I think it worth adding the separation of visual and computer readable is important.

[juliusgb]

Yes, exactly.

Conceptually they're distinct now (even in our minds). Sometimes they intertwine. The Miro board (pinned on the discord channel) has more details.

As for the setup on github, an initial idea was to have all in one repo (#5) before moving each to its own repo.