Mapmakers: The Digital Life Collective
In the Mapmakers series, we make design changes to Kumu maps, working our way toward deeper insight by applying (or intentionally not applying) different Kumu features.
The Digital Life Collective is a co-operative (a member-owned, member-funded, and member-controlled organization) that researches, develops, funds and supports “Tech We Trust”—a phrase they coined to describe technologies that prioritize our autonomy, privacy and dignity.
If you’re geeking out after reading that, you’re not alone! I think this co-operative is a super cool idea, and I’m thrilled to tell you that its members are also Kumu users ⭐ 🚀.
The Collective used Kumu to build a prototype network map to explore the ecosystem of what might be trustworthy technologies. In the map, most of the elements represent technology projects, but there are also elements that represent organizations, like Mozilla, that are working to preserve and promote a free and open internet.
You can also click on elements in this map to bring up a profile with more information. Here’s what the profile looks like for Loomio, a collective decision-making software that another Kumu user (Richard D. Bartlett) helped build:
The Collective used decorations to color-code technologies and organizations, and they put a legend in the bottom left corner to help the readers understand the map.
The colors they used are pretty nice — not too bright, and not too bland — but I noticed that they were picked from Kumu’s standard set of colors. I poked around on the beautiful, free, open internet (let’s please keep it that way!), and I found that The Collective has some snazzy brand colors. So, to give the map a more unique and brand-aligned feel, I used decorations to apply those colors instead:
These new colors (“Persian Green” and “Deep Lemon”, according to the Collective’s branding guide) also have a wider difference in their hue and lightness than the original green and blue. It’s always good to aim for this kind of difference when you can—if anyone with color vision deficiency is reading your map, they’ll have much less trouble distinguishing between your colors!
Now, you might have noticed in some of the screenshots above that there’s a small bit of text at the top that says “Cluster by:”. This is the label for a cluster control that the Collective built to allow map readers to activate clustering for several different fields at the click of a button.
Round of applause for the Collective, ladies and gentlemen 👏👏! Controls are one of Kumu’s most powerful features, allowing you to customize how people interact with your map and make the data come alive.
The Collective’s cluster control gives you a number of different options to connect organizations and projects based on the data stored in their fields. Clustering by maturity, for example, allows you to quickly scan this map to find projects in the alpha stage, beta stage, and beyond:
Some of those options, when combined, turn up really interesting comparisons, like this map of projects clustered by 1) the organization that maintains them and 2) the other projects they rely on.
But, some of the other clustering options produce views that are either too simple or too dense to really highlight key insights from the data.
Clustering projects by Level of Diligence produces a straightforward hub-and-spoke effect:
Clustering projects by Purpose produces a hairball in the middle of the map:
How to remove the “hub-and-spoke” effect
When you notice that clustering on a certain field only gives you a simple hub-and-spoke diagram, you’ll want to take a closer look at that underlying field. More than likely, there will be little or no variation in field values. To get more insight from this kind of data, you can try two solutions:
- Instead of using fields and clustering to make the data visible, simply use the map overview to give more context about your dataset. In this map’s overview, for example, we could add the sentence “This map shows projects whose level of diligence is apparent, based on their stated purpose.” This approach will allow readers to focus on other, more insightful views in the map, without completely forgetting your original data.
- Or, you can try to replace your field’s predominant value with a more varied set of values. In this map, for example, we could comb through the data again to replace “Apparent” (present in almost every project’s Level of Diligence field) with one of three scaled values: “Not very apparent”, “Somewhat apparent”, and “Extremely apparent”. When you cluster on those varied values, you’ll almost certainly create a view that offers more insight into the complex data on the map.
The first solution is quicker—adding a single sentence to the map overview is much easier than reviewing the field values of every single element—but if the field you’re reviewing is really important for the map, the second solution might be well worth your time.
How to remove the “hairball” effect
We have a love/hate relationship with hairballs! On one hand, they’re usually a sure sign that lots of interesting relationships exist in the map. On the other hand, they’re rarely legible.
Clustering is likely to create a hairball when:
- The field you’re clustering on has more than 2 and less than 10 options,
- Each element can store multiple options in that field, and
- You’re clustering 25+ elements by that field
Every map is different, so these conditions won’t always create a hairball, and you can definitely achieve one under other conditions. But, regardless of how you achieve the hairball, Kumu offers some reliable solutions.
First, you can try turning on the “hairball” layout preset. This will spread out the elements and stretch out the connections, and that extra breathing room often makes the map much easier to read, and therefore much easier to draw insight from.
The screenshot of the Collective’s map above already has the hairball preset turned on, so we’ll test out another good solution: flags.
When you decorate elements with flags, Kumu draws colored arcs around the outside of those elements. Each flag represents one of the possible values in your field.
Flags are a clean, colorful, readable alternative to clustering when you are working with a multi-pick field that has about 15 possible values or fewer. (When you exceed 15 distinct colors, it’s starts to get pretty difficult to distinguish them from a distance!) Here’s the the Collective map, using flags instead of clustering to visualize the data in the Purpose field (flag color legend in the bottom right)
After you add flags, we recommend adding a filter control so readers can filter by those possible values, and your map will get even easier to read. Here’s a view of the flags and a filter control working together on the Purpose field:
This view filters all the projects to show only the projects whose purpose includes digital inclusion & equality. Comparing this view to the original above, you can see that almost all the projects have been filtered out. Of the ones that remain, almost all are involved in either decentralization or privacy, in addition to inclusion & equality.
The insight we get from this view: if we want the internet to remain free, open, and in the public interest, we need more trustworthy technology projects to prioritize inclusion & equality, not just the well-trodden ground of privacy and decentralization.
Admittedly, readers can probably glean that insight from the original view. But, in our experience, the flags + filter view will be much easier to digest than the cluster view, especially for readers who are new to network viz in general.
That’s it for this map! Kudos to Kumu expert Christina Bowen for assembling the data on behalf of the Collective, and of course, props to the Collective for tackling a complex problem on behalf of all of us who use the World Wide Web. The map we reviewed above is just their prototype version—if you’d like to help them create the final version, or want to see your organization or project represented on the map, definitely check out their ecosystem directory and fill out their ecosystem survey.
Want to submit your map for review? Send it to mapmakers@kumu.io!
