Modern GIS – A New Approach to GIS
Our guest this week is Matt Forrest, the VP of Solutions Engineering at Carto. Matt began his journey in the geospatial space back in college when he decided to major in Geography. After college, he continued to enjoy doing visualizations, cartography, and analysis. His career later on took a new turn when he started to engage with Carto in 2012, and eventually joined them in 2015. Since then, he has been using a mix of different geospatial technologies in different creative ways to solve problems. Matt provides us with an interesting definition of how we can view Modern GIS:
The processes, systems, and technologies used to derive insights from geospatial data. Modern GIS uses open, interoperable, and standards-based technology. It can be run locally or in the cloud and can scale to work with many different types, velocities, and scales of data.
Have We Been Doing GIS Wrong?
The idea of Modern GIS may trigger thoughts that this is the only right way of doing GIS, and that we have actually been doing it wrong all along. That’s not quite the case. The term ‘modern’ should not be considered interchangeable with ‘right’. To clear any confusions that may arise, Modern GIS does not mean right as in it is the only way of doing things. Rather, it is an outlook of what Modern GIS might actually look like. It is critical to have it at the back of the mind that it is in no way meaning the best, or the only way.
The Key Principles of Modern GIS
Some of the main principles that we can pull from the definition of Modern GIS include: Open Technology, Interoperability, and Scalability.
Interoperability fosters collaboration among professionals. It makes working together easier, and more efficient. Professionals can connect and create different outputs from different places, as well as connect different technologies they are using in a project.
One remarkable feature of Modern GIS is scalability. It opens such a powerful window that makes it possible to go from one computer to serverless processes, with hundreds of nodes at a massive scale. The relief here is that anybody can run it – it does not matter whether you are on your machine with no internet connectivity, or running it on massive servers on the cloud. Modern GIS works on any of these scales.
Local vs. Cloud?
In the current geospatial atmosphere, there is a lot of activity surrounding the cloud, and a lot of things are moving towards the cloud. Still, Modern GIS allows those same technologies to be run locally. It is freeing to know that you can run this on your machine on-premises if you don’t have cloud resources available.
Modern GIS does not imply that you have to use the cloud. You don’t have to since it can be run locally. The choice is yours to make depending on your needs. It is not necessarily a local vs cloud thing, although there are a lot of advantages to using the cloud.
When core concepts are out in the open, people work with them across technologies, and across teams. It makes it possible to easily move between databases and toolkits when the languages are common (i.e. Python, SQL). This enables a connection to systems that are largely open and scalable. In the future, it is possible that the broader geospatial technology industry is leaning towards the open-source solutions.
If you wonder just how critical open-source technology is, then you may be amazed to learn that even at the core of proprietary technology, open-source technology is used. In the words of Ramsey;
From a business point of view, the cloud wouldn’t work without open-source technology. Think of it, if one had to pay for every instance of running Geode, running it on a massive scale would be unsustainably expensive. Without using foundational open-source tools at their core, it becomes very hard for proprietary technologies to succeed and grow.
The Advantages of Modern GIS
There are many possibilities as to the types of outputs in Modern GIS – and there is no wrong form of deliverable. So many different outcomes can be developed, which adds flexibility in creating what is needed to solve a problem. The solution may be a dashboard, API, micro service, mobile app, data pipeline, data collection tool, machine learning model, static map, report, or any combination of these.
The bottom line is that with Modern GIS, you can walk away with the tools to build what you need for solving a problem, no matter what it is.
You can also connect to non-geospatial APIs and input information which can add valuable context to help solve a problem. In Modern GIS, you are not constrained on what you can deliver. You can actually come up with something completely new, and build it. There is no requirement that the output always be a map.
Are static maps going away?
Geospatial is naturally visual – and that is certainly not going to change. Visualisation is very important to the way we communicate geospatial data and analyses. Despite this being unlikely to change, we can start to look into what are some of the other pieces that are happening in the visualization. The traditional map is a great jumping off point because people naturally understand them.
Breaking Out of the Geospatial Silo
Modern GIS breaks out of the traditional silo of GIS, where the technology exists only in the context of its own world of tools and systems. To the contrary, now GIS is an interoperable asset of your tech team that can be leveraged directly alongside other technologies, using the same languages and processes that they already use – and you do not have to be an expert on everything. You can learn just enough parts of the other technology to help you integrate what you need.
Access to All
Modern GIS promotes access to all, and this includes accessing the power in low-code environments. Despite this, it is still good to know some code. The most important thing is to learn and understand the concepts, and maintain the choice to go deeper if you want. Low-Code/No-Code is on the way, but if you want to get specific, you will probably still have to do it yourself.
The (Modern) GIS Learning Path Dilemma
Anyone anywhere can start practicing Modern GIS, but the challenge is in understanding the learning paths of how to do so. Here, the geospatial space faces a great dilemma. There are no set learning paths that are really viable to help people grow. There is no clear definition that informs people of the exact outcome when they choose a particular line of learning. Even in real world positions, there is always a mix of skills that one is expected to have.
Currently, many new paths are coming up. One of the best things about Modern GIS is that it will support new job functions and new roles. It has already started to grow, and careers with more defined learning paths are starting to appear. For instance, due to the explosion of data in our world today, Geospatial Data Engineering is gaining ground fast. This career puts together pieces from Data Engineering and Geospatial technology, making it closely relevant to the modern problems various industries face today.
Tips on How to Start with Modern GIS
When thinking of adopting Modern GIS, there are so many things to learn that it may seem daunting. Here is some advice – start with just one thing. Ultimately, you have to start somewhere, so why not the beginning? First pick up foundational pieces that you can build on later. Understanding the core ideas really make it easier for things to link together as you progress.
A good place to begin may be from something in your job, or a favourite project. This could be a problem or something that is difficult, like a new responsibility or task in your current job. Then try to find a way of solving it using just one tool, like Python.
The other piece is to find a community and network. There is no right path, but being in the same pool with great minds presents an opportunity to learn how other people got their careers, and can inspire you to tackle the next big thing. With this information it is possible to develop some kind of path that can make learning a lot easier.
A Learning Approach to Modern GIS: Walk, Run, And Sprint Process
It is important to start simple. You may run into issues along the way but feel prepared and know that it is okay for mistakes to happen. The walk, run and sprint process is a breakdown of how you may break into Modern GIS:
WALK – Learn QGIS, then Python and Geopandas. These are some of the core tools that data scientists use, so they will give you a strong foundation early on. Although a desktop product (QGIS) might sound a bit old-fashioned, it serves as a good entry point. It is great intro, especially if you are coming from other desktop GIS software like ArcGIS.
RUN – Learn SQL at this stage. Understanding it enables you to take a lot of processes to the database and this makes the much faster as they are closer to the data.
SPRINT –Eventually, learn how to scale things up to a massive scale on the cloud.
Currently, there is a huge pool of tutorials and guides (including right here on MapScaping) that make it less daunting to learn these tasks, unlike several years ago when resources were not widely accessible. There is also room for collaboration and learning across different geographies, locations, and educational backgrounds.
Modern GIS is indeed a new approach to doing GIS. It is a call to rethink our positions as GIS professionals. It is the perfect time to embrace the idea that things could be different, and help people know what else we can do. We can become almost entrepreneurial in our jobs in the sense of seeking out opportunities to do interesting work. There are unlimited possibilities on how we can derive insights from geospatial data to tell a story, solve a problem, and help someone.