Look I have been talking about modern GIS for about three years now (a full three if you count this post). And the adoption of the term has grown significantly since my first full article on the topic in January 2022:

But in the past year or so something caught my eye from a specific company that says across multiple content pieces that says “Modern GIS is Web GIS”.

Long story short, no. But let’s break it into two groups of two key issues each:

As it relates to cloud-native design principles:

1. The principles of cloud native computing and modern data stack
2. The principles that define modern GIS

And as it relates to services labeled as modern GIS:

1. Wrapping traditional GIS in containers doesn’t make it modern
2. Web GIS is just visualization that happens to live on the cloud

Cloud-native and modern data stack principles inform modern GIS

Before we begin we first have to agree that GIS is a set of technologies and not an area of study or practice. I have discussed this here in more detail so I will let you read that if you want to go deeper.

Now what makes modern GIS, well modern, is that it adheres to the principles of modern cloud computing. Namely two well defined areas: the modern data stack and cloud-native computing. Let’s review the high level principles of each of these.

Modern data stack

The modern data stack offers a new architecture for data and analytics that looks to be more modular and innovative, leveraging cloud technologies to ensure that services can scale quickly and easily. It also focuses on modularity, meaning the ability to work with specific tools that focuse on one specific part of the problem, and interoperability, or the connectivity between those modular tools. Let’s take a look at some of the core principles highlighted from two different sources.

This article from ThoughtSpot focuses in on the following elements:

• “Easy to try and easy to deploy” – i.e. SaaS-based, user-friendly, integrated, and easy to deploy
• “Massive scalability” – i.e. Tools offer massive scalability for data, users, and a range of use cases.
• “Velocity” – Scaling growth brings data complexity, and modernizing platforms early prevents technical debt and enables efficient, data-driven operations.
• “Innovation through agility” – Innovative companies thrive by quickly validating ideas, iterating, and treating data as a dynamic, evolving asset.
• “Composable data stack” – Treats each tool as a configurable component within a larger architecture.
• “Greater efficiency” – The modern data stack prioritizes usability, efficiency, agility, democratization, and reduced manual management.

In short, smaller more scalable component enable the ability to grow and be more nimble.

Now let’s take a look at some additional points from this article from The New Stack. It includes some other similar points but these are the unique points:

• Cloud native: Provides scalability and flexibility across public and private clouds, supporting multicloud compatibility to prevent vendor lock-in.
• Performant: Prioritizes high performance, enabling efficient data processing and analysis.
• RESTful API: Ensures smooth, standardized communication between components, promoting interoperability and microservice creation, exemplified by the S3 API.
• Decoupled compute: Separates compute from storage, allowing independent scaling for optimized costs and dynamic resource allocation.
• Open: Embraces open-source solutions and open table formats, encouraging collaboration, innovation, and adaptability across platforms.

With these additions, we add in additional context around cloud-native, debug-coupled computing, and open and interoperable systems. These are key things I want you to keep in mind when we continue to look forward into the assertion that modern GIS is web GIS.

Cloud-native architecture

As mentioned above there is also the cloud-native archtiecture that is a foundation of the modern data stack as well. Part of that is understanding the parts of a cloud-native architecture but also the variety of cloud deployments that exist. Let’s take a look at the first using definitions from this article from DataStax:

• Cloud enabled: Legacy applications migrated to infrastructure as a services (IaaS) for faster updates, but retain rigid architectures that limit cloud scalability.
• Cloud based: Redesigned to run efficiently in the cloud, supporting automation and flexible pay-as-you-go scaling.
• Cloud native: Built exclusively for the cloud, leveraging full scalability, security, and cost efficiency with cloud provider support.

Within that, cloud-native architecture applies to many different applications and offers a core set of benefits. A cloud-native architecture offers high availability, automatic scalability, enhanced monitoring, faster development, and built-in security. It empowers teams to develop and scale applications efficiently while keeping them secure and responsive to user demand.

In short and summarize by this article from Okta, the benefits are speed, low cost, and options.

The elements that make up a cloud-native architecture get more into the infrastructure layers: micro-services, optimized containers, services meshes, immutable infrastructure, and cloud-native APIs. If you don’t know all of those things, don’t worry you don’t need to – I had to look up what a service mesh was if I am being honest.

While this is likely less relevant from a user’s perspective, the various definitions of cloud enabled, based, and native will be since that helps define if a technology is simply moving to the cloud without innovating, or if it is positioning itself to be truly scalable as a part of a modern data stack, or in our case the geospatial data stack.

What is modern GIS then?

When I wrote that blog post about modern GIS in 2022 I set out a specific scope of key elements that helped to define it, which are anchored in the two architectural elements outlined above. These are not approaches or techniques, but core technical foundations that will help a technology like GIS scale and become more impactful. Take a look at the chart below:

Now let’s label this with elements that are from the modern data stack (MDS) or cloud-native (CN) architectures.

Many of these concepts originated within the modern data stack but they apply to the trends taking place in geospatial data and analytics today, particularly with the trend of new tools focusing on one specific area rather than trying to be a self contained monolith.

So why say that modern GIS is web GIS?

I think there are a few reasons for saying that but first let’s try and define web GIS. I ask my friend ChatGPT what it thinks:

Web GIS (Geographic Information Systems) is a platform that allows users to view, analyze, and interact with geospatial data through a web interface, without needing specialized desktop software. Essentially, it brings the power of GIS to the internet, enabling broader accessibility, collaboration, and real-time updates.

In my mind the whole thing comes down to this sentence: without needing specialized desktop software. GIS has for so long revolved around the desktop GIS toolkit. It was the single entry point and tool for, well, everything. Cartography, data management, data transformation, analytics, spatial relationships, visualization, and nearly every thing under the sun has a place or a tool in a desktop GIS.

For what it’s worth I think that desktop GIS has a place in modern GIS and can still be a tool to access different data tools and all the other parts of the geospatial data stack.

My issue with this statement is that it implies that modern GIS is just the simple process of moving your GIS to a web based tool. It implies that you just lift and shift to a web GIS, but all those same workflows are still contained in a single tool rather than leveraging the right tool for the job and working in a composable data stack.

The other piece is that while some tools and services have been adapted for the cloud, the services themselves are not scalable so they fall into the category of cloud-enabled or cloud-based at best.

In short, if a service you are using is not capable of scaling in a current set up, moving something to the cloud doesn’t magically make it more scalable. Map tiles are a good example of this. If you are having trouble generating map tiles at scale, throwing more computing power might help, but it is not going to make it suddenly more efficient than it was locally.

Or, a fresh coat of paint wont change the foundation.

But wait a minute…

Before we go I do want to draw your attention to this document, and the main points in it.

Data as services not shapefiles

Discover, publish, share, and manage real-time, structured, or unstructured data—in spreadsheets, CSV files, or cloud storage—as web layers and services. Students can move beyond shapefiles to understand how GIS is powered by dynamic, interconnected services.

Look anything to get beyond shapefiles is good in my book, and the fact that we are talking about data in cloud storage, well that is a good start.

Configurable web applications not map layouts

Build configurable, responsive web and mobile applications. Move beyond the static maps that have been the heart of GIS for decades. Modern GIS is Web GIS, powered by lightweight apps that meet the needs of a digital-native audience.

This has the offending line in the quote, but I think the important part is what is right after it: lightweight apps. In the geospatial data stack that layer can be anything: a GIS tool be it cloud or desktop, an analytics toolkit, Python and embedded data science tools, and even applications.

Automation and scripting

Publish data, automate workflows, provide notifications, update and synchronize data, and more. These skills have always been part of GIS, but they have become even more critical in the fast-paced data and information space.

Automation? This is the backbone of the modern data stack and in turn the geospatial data stack. The ability to leverage transformation and automation tools at any layer is critical to remove manual effort from the process and serve geospatial insights without human intervention.

Cloud infrastructure and architecture

Teach students about web architecture and the Web GIS information model. Students need to know the underpinnings of modern GIS. It’s critical to understand and troubleshoot issues in cloud computing platforms and databases to keep systems running.

Cloud is a core element of the modern data stack and ensuring that you have the ability to not only work in the cloud, but automate processes and create a data stack that works for your specific issues is key. So this one makes sense as well.

While some of the tools and methods that are being suggested in the doc to implement these things are not totally in line with the ideas of modern GIS, it is clear that these foundations are stating to take hold at all levels of the geospatial ecosystem. If you told me that ArcGIS Pro would support DuckDB, I would have been quite surprised, but in fact that is a reality and in my opinion, a good thing.

---

If you made it this far, first of thank you, and I wanted to share a project I have been working on. I have done a ton of writing about modern GIS and shared a variety of tutorials across the web in video and text form, and I even organized an online community to change the way we learn modern GIS. But I never organized that information into a course or learning platform to enable anyone to learn and ultimately become certified in these concepts.

So with that goal in mind I just launched Modern GIS, a place to learn about modern GIS, gain new skills, and earn certifications. The first course, Foundations of Modern GIS, covers these exact concepts discussed today and as far as I know is the only course that provides a way to be certified in the foundations of modern GIS. I would love for you to check it out, and for the readers of this newsletter, you can get half off using the code SPATIALSTACK50.