Don't know if you spotted this - Excellent graphic vocabulary for representing relations. Grist to your mill maybe, in graphical representation of value networks?

@mike_hales I had not spotted it before. Very timely, will bookmark and study and pass on to some new devs who are working on that kind of thing. Here's one of their examples, with the code to generate it:

@mike_hales that link gave me a spinner for so long that I gave up. Did Ward intend to implement some of that stuff in his fedwiki?

In response to your Valueflows chart he posted
> I think we could draw this picture using our algorithmic graphviz except possibly for the clusters where we have trouble making more than one for reasons that I can't remember. We would make each box or circle be a page and add "qualified" links for the connections.

@bhaugen 2of2
In response I posted
> Making each operation or stage a page would be fully consistent with the language used to generate these graphs. The graph is a 'valueflow', specified using the Valueflows description language.

> A page per stage is natural - parameters of that stage can be tweaked in that page. Computations can run along the network, on numerical parameters managed in the stage-page.
Am I right
bhaugen: ?

I guess you didn't get the notification?

> I guess you didn't get the notification?

No, I didn't get to that matrix link at all. Might try again another day on another browser after rebooting. I'm interested.

And thanks for being the relay...

Re your substantive questions, I'll need to re-study that graph focusing on clusters.

But one of the devs working on a related project has a mockup of a simpler, less-diagrammatic network flow textual data management user interface that is sorta like Fedwiki: side-scrolling thru process stages. If she gets it implemented, I'll yell.

It's for this network, that we are working with now. Don't know if I showed you this before:

> Re your substantive questions, I'll need to re-study that graph focusing on clusters.

Clusters = subgraphs. Had to read some of the code to learn that.

So Ward's comment about clusters is a code-level problem, not a resource-flow-level problem.

> Am I right

You are right. More Sunday latest. Busy days this last week...

Got there, finally, and will continue there later today. Studying some details...

Sign in to participate in the conversation

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!