As CartoDB rebrands to CARTO to reflect democratization of its technology, CEO Javier de la Torre reveals how open data can be used for innovation and economic growth
CARTO began as an open-source product. What was the importance of being open source for you?
Upon a closer look at the state of the industry, you will see a lot of innovation is happening in the open source and open data space. Embracing this movement essentially provided us with:
• Development speed: We stand on the shoulders of giants. Leveraging and collaborating with this community has definitely allowed us to get to market faster.
• Access to talent: As this industry is getting hotter and hotter, there is a huge demand for geospatial experts. Being open source gives us access to some of the most talented developers.
• Transparency for clients: In an industry dominated by highly proprietary incumbent companies, being open source definitely feels like a breath of fresh air to our customers. And as value further increases with the move from licenses to service (with SaaS), there is no significant impact on the business side. The transparency that open source provides gives our customers confidence in terms of avoiding ‘lock-ins’ and extensibility capabilities.
Overall, practicing an open source philosophy when it comes to our software development has provided us with much more value than we could have obtained by making our software proprietary, and it has opened collaboration doors with some very large companies along the way.
Your rebranded company comes with a new tagline: Predict Through Location. How would you summarize the power of location?
Location intelligence is part of a mega trend around data-driven processes. This data is enabling automation software, and this is changing not only the Internet, but also the entire world. Every process that can be automated and replaced by software, will be, and they will all need data to operate. When you think about the fact that everything happens somewhere, you realize that location is going to be a fundamental part of the information needed to power these automations. So, as part of a bigger trend, location intelligence, through machine learning, will enable not only self-driving cars, but many other things in our daily lives that will change society in a profound way.
If apps would have to pay every time they acquired a GPS signal, we probably would not have an Uber or a Pokemon Go
Is the location component becoming more powerful with increasing focus on the Internet of Things?
Definitely. Location sensorization in IoT and mobile are generating new, massive datasets that open up entirely new opportunities. To realize the value of understanding processes through location, you need a significant amount of data and coverage. IoT is providing just that by enabling the value of location intelligence to flourish. Not only is the location dimension becoming more real with more data, but it is also driving a huge amount of research and innovation, which will ultimately make it much more powerful.
Do we need to democratize location data by making it openly and freely available?
Parts of it for sure; other parts will need to remain closed, and that’s acceptable. For example, geocoding must be more open in order to reach democratization. For many use cases, the cost of geocoding data makes it too expensive to even try some spatial analysis. Think of it like the GPS signal. If apps would have to pay every time they acquired a GPS signal, we probably would not have an Uber or a Pokemon Go. Some core datasets work as an index for many other things that require it in order to be freely available.
Another aspect of opening these datasets is around transparency and quality control. If you look at services that are providing customer segmentation information, you will get very different responses from each service. Each of them act as a black box. Which one should you trust? This will be a bigger and bigger problem as these services become a foundational building block for other analysis built on top of them. This is also an area of ‘Derivative datasets’ that need to be open. That is in part the work we are doing with our Data Observatory.
How do you think open data can be used for innovation and economic growth?
We have already seen several models in place. For example, with our Data Observatory we are collecting open data around demographics and census information from many different countries. We make use of that data by creating segmentation layers and other derivative datasets. CARTO takes care of the processing and packaging of these datasets to make them readily and easily available to those users who typically would not have been able to take advantage of it. We have seen cases where we have reduced the operational costs for some organizations by millions of dollars by using better information.
The other aspect is the one that I mentioned previously. Imagine, there was no cost associated with geocoding information — if all zip codes, street addresses, etc., were publicly available. Companies like CARTO would aim to make them available immediately with services that the masses could leverage. This means more spatial analysis could be done on this data and many more insights could be found. Again, this is a great example of how opening data can foster innovation and consequently economic growth.
Do you think open data has the potential to break down information gaps across industries?
We have not yet seen many examples of private organizations opening data to improve efficiencies within their industry. There is certainly great potential in this area. Think of industries like car manufacturers in the context of the current revolution of self-driving cars. In order to enable them, we will need lot of data that will, in part, be collected through the sensors of the car itself. Open data can provide a clearing mechanism for sharing this information between manufacturers and other industries to ensure companies are competing against a level of standardization. This could also make insurance companies run more efficiently, and the same goes with data generated from taxi services.
We are likely going to see consortiums in the future between the private and public sectors where they use open data as a clearing house mechanism to collaborate.
Do companies lose competitive advantage when they share data?
It all depends on the data and their business model. In some cases, they do and in some cases, they don’t. What happens often is that companies play the ‘secure’ card by declaring all data ‘non-open’, just in case. This is something that is likely to change in the upcoming years. As organizations start thinking deeply on their data’s lifecycle, they will likely identify those cases where opening data provided worthwhile advantages and when it is better to keep it closed.