We have made a new OSM-Carto release that finally includes a solution for one of our oldest issues – the access restriction rendering on roads. And i want to make a few additional comments here beyond what i wrote in the release announcement.

The issue in question had been open for 11 years and i already presented the timeline of that process in the development discussion:

• A few months after the issue was opened a first selective ad-hoc change for slightly extended access tag interpretation was rejected and the decision made that we need to have an overall strategy on this topic (semantics, rendering) rather than to make ad-hoc changes.
• It took nearly 5 years after that for a first concrete proposal for a tag interpretation strategy to be presented.
• Another 2.5 years until a first proof-of-concept to address the matter within the technical and design framework of OSM-Carto – my work presented on this blog.
• Another 2 years until an OSM-Carto developer decided to pick this up and develop a concrete solution for OSM-Carto.
• Finally: half a year of daunting work in trying to get to an agreement among the maintainers on the specifics of the implementation.

So what can we learn from this and why did it take that long?

What takes time in map design?

First this timeline as i see it approximately reflects the relative amounts of work required for different tasks in practical map design that tries to be innovative and not just copies ideas and methods developed elsewhere:

1. 50 percent of work is conceptual design work – a viable and well thought through principal idea how to show something and a conceptual sketch of how the rules need to look like to get from the data to that visualization is already half the way.
2. 25 percent of the work is developing an initial concrete implementation of the conceptual design. Depending on the problem this consists mostly of algorithm development, symbol and color design.
3. another 25 percent of the work is spend on refining and adjusting this initial implementation to the practical constraints of the concrete map. This includes technical optimization work, design adjustments based on user feedback and practical testing of the design in its cartographic context.

These percentages are under the assumption that the conceptual design work is of high quality. If it is not you will often end up not being able to finish the second step and need to go back and repeat the conceptual design work with the lessons learned from the unsuccessful first attempt. This is normal, even for experienced map designers. But it makes the whole process take much longer of course and you still end up with similar relative percentages.

The problem is of course that this is hard to sell. And this is not much easier in a volunteer community project like OSM-Carto than in a commercial project. I can’t count how many times i have seen comments on OSM-Carto of someone essentially saying: Just render it in some form, does not matter how.

So to be clear on that: Yes, you can follow that approach. You can essentially cut step 1, reduce step 2 to copying things from elsewhere and making random choices in colors. You will end up with something you might feel fine calling a map. But that is essentially just map design parasitism. Economically, this might even be lucrative because of the time saved. But intellectually and artistically this would be highly dissatisfying. Which is why i don’t think this would be viable as a guiding principle for a volunteer community project.

Attracting talented and skilled contributors

But all of that of course does not explain why it took 11 years for OSM-Carto to address this specific issue. In particular in light of the fact that, in contrast to other problems, this one was technically not particularly challenging.

The vast majority of these 11 years was of course not spend on actual work, most of this was essentially waiting for someone with the necessary skills and the interest in the project finding the time to work on this.

The bottom line is: The resources bottleneck of a community map design project like OSM-Carto is qualified map designers with the capacity and the interest to invest their time into the project. OSM-Carto has been better at this than many other map design projects in the OSM-Community because being the main map for the OSM-Community is a huge incentive for contributors.

But – as i have pointed out in the past – we, the OSM-Carto maintainers, have over many years failed to provide talented and skilled map designers an attractive environment to contribute. The ability to do conceptual design work on concrete cartographic problems depends on there being a clear overall strategy for the map as a whole and we have not been able to agree on such a strategy for OSM-Carto for a long time now.

The other factor is the social context of the project. How non-technical work (which – as outlined above – is more than half of map design) is valued and appreciated in the larger OSM-Community has a huge effect on how much talent and skill in map design OpenStreetMap can muster overall. I have pointed out in the past as well that the OSM-Community is severely lacking here. We have a lot of contributors who show significant latent interest and often also talent in that field but do not find a nurturing environment to develop those.

Conclusions

The bottom line is: There is immense (i would go as far as saying globally unique) potential in the OSM community for talent, skill and interest in high quality map design with a broad range of cultural backgrounds. That potential is currently largely unused because:

1. Community map design in OpenStreetMap, above all OSM-Carto, lacks the ability (in terms of the human ressources capacity, in terms of the technical tools and – in case of OSM-Carto – also in the form of a common strategy and vision within the project) to provide those people the supportive environment they need to develop their abilities.
2. The OSM community collectively under-values map design and under-estimates the significance of the ability of OpenStreetMap to be innovative and actively shape the state of the art in that field for the long term future of the project. As a result people do not have the impression that investing in community map design in OpenStreetMap is worthwhile either socially or in terms of their own cartographic ambitions.