In this episode, Marco Bernasocchi, co-founder and CEO of OpenGIS.ch, introduces us to QField, an open-source mobile application designed for field data collection in conjunction with QGIS.
Marco shares his journey in developing QField and discusses its seamless in integration with QGIS, allowing users to capture, survey, and manage geospatial data on various mobile devices.
We also discuss the technical aspects of QField, including its user-friendly interface, the ability to connect with external sensors, and the recent introduction of QFieldCloud for enhanced data synchronization and management.
Marco highlights the application’s diverse use cases, from citizen science initiatives to archaeological documentation and utility inspections, demonstrating its potential to transform data collection processes across various industries.
More information on Qfield:
Or https://www.opengis.ch/#contact
On a personal note, I have been working as a freelance Geospatial consultant for some time now and one of my projects is slowly winding down, which is why I am looking for new projects to get involved in!
If you need expertise in Geospatial consultancy, GIS management or the marketing of geospatial products and services Please reach out!
https://www.linkedin.com/in/danielodonohue/
or contact me here info@mapscpaing.com
Full Transcript
Hi Marco, welcome to the podcast. You are the co-founder and CEO of OpenGIS. ch, and you describe yourself as being the father of Qfield. We’ll get into that in just a second, but let’s start with a brief interaction. I’m sure you can put more words around this than what I just did. Who are you, and what is OpenGIS. ch? Yeah, as you said, I’m Marco Bernasconi.
I founded OpenGIS 12 years ago with the idea of pushing open source GIS software further, and to make a living out of it. Meanwhile, we are 22 people working, mainly in Switzerland, that’s the . cih part of the name, and do a lot of work in QGIS itself, infrastructure around it, web infrastructure, and of course everything that has to do with Kewfield. Okay, cool. So it sounds like you’re almost running a consultancy.
What is Qfield? How does that fit into your consultancy, your business? We do a lot of consultants as well, but we do see ourselves a lot as very technical people, so we do a lot of development. We have five or six core developers of QGIS itself, so pretty technical, pretty developer-heavy, and as I said, Qfield is one of our three main kind of branches within the company, and it is this fantastic application for managing your data on the field, capturing, surveying, editing, and so on of whatever data is important to you. Okay, how does this fit in with QGIS? You might have just mentioned it then, but I didn’t fully understand it. You can imagine Kewfield as being a QGIS optimized for the field.
In fact, there is a full QGIS running in the back of a very intuitive and very simple user interface, and that’s the relationship we have to the QGIS project. It’s a very mutual kind of thing where sometimes when we need new functionalities in Kewfield, we actually build them in QGIS because we get them back in automatically. That was the next question I wanted to ask.
What does this mean? Does this mean when new updates or new features become available in QGIS, are they then available in Kewfield? Yes, not forcibly automatically. So if you imagine a new feature in the user interface of QGIS will obviously not come to Kewfield directly because we don’t use the user interface of QGIS itself. We build our own interface. But if you have a new symbology renderer or if you have something new in the Atlas printing or in the reports or in various other parts of the code, then we get them directly back in Kewfield itself. For example, if you remember a couple of versions ago of QGIS, we needed to have the possibility to have sensors sending data into Kewfield.
It was something that was very interesting for QGIS itself. So we built it in QGIS and then we got it back into Kewfield. A lot of things happened this way in QGIS. I really want to come back to that sensors later on because I have a few questions around that. But also about the interface, because to me, you say you’ve got a full version of QGIS running on this mobile platform. How do you deal with that in terms of the interface? There must be a need to limit what people can do.
I mean, typically mobile devices aren’t as big as desktops. Yeah. So if we go a little bit back in the history of Kewfield itself, the very first version of Kewfield that was running was called QGIS for Android and I built it in 2011, I think. And the idea then was really just to make QGIS run on the mobile device. And I promise you, I had a full QGIS user interface running on a three-inch screen. It was the most awful user experience you could ever imagine, but it was the beginning of a very great plan that I had back then already. The idea had already been, first I make it work, then I make it look good, and then I make it integrate.
And we currently are really, we reached those three steps. So the main point you asked here is the making it look good. The user interface of Kewfield is a complete separate thing from what QGIS user interface is. We are not reusing anything from the interface of QGIS. Basically, we are using all the background libraries of QGIS. QGIS is luckily built up in an almost modular way where you have QGIS Core, QGIS Analysis, and so on and so on, QGIS UI. And we’re just completely dropping the UI and now rewriting a complete new user interface to make it adapted to people in the field.
Because as you say, screens are small, screen state is very precious, and QGIS has a lot of buttons. So that’s why we took this path of completely rewriting the user interface on top of QGIS Core and QGIS Analysis basically. Before we carry on and talk about what a possible workflow might look like, I’ve just got a few questions that I think is probably worth answering now. When we say mobile device, do we mean just Apple, just Android? Are we talking about all mobile devices, phones, tablets?
What do we mean? QField runs on basically anything. We are building a matrix of five versions. So we’re running on iOS, on Android, as you say. It runs on Windows, it runs on macOS, it runs on Linux. And why do I mention that is because QField is not only for mobile devices, it’s actually for any devices. For instance, we’ve seen universities starting using it as introductory GIS system because QGIS has so many buttons that it’s a bit scary for a beginner.
And so you can have QField running on your desktop as well as a kind of a very simple, intuitive user interface to a QGIS project without having to create a new project because QField configures itself based on a standard QGIS project. So any mobile device you want, tablets, phones, whatever, you can use it on three-inch screens again now, and it still looks much nicer than it used to with the original version of QGIS on Android. And then you can go up all the way to Microsoft Surface devices. For example, there are plenty of companies that already have Surface laptops, tablets in the company, and they don’t need to get another tablet. They just use the Microsoft Windows version and go out with those tablets. You talk about it being almost like a simplified version of QGIS. How would you describe this then?
Is QField, is it a data viewer? Is it a survey tool? Or is it also a data processing tool in the same way that we think of QGIS as a data processing tool or an analysis tool? The main focus of QField is for sure data capturing. So it’s really go out and capture data, go take pictures, take videos, take audios of your assets that you have out there, modify attributes, go and check if your fiber-to-the-ohm box is fixed and report back to your centralized database that has been fixed, report back damages, pictures of natural catastrophes. So those kind of things is really where we put the main focus on. But it can also just be used as a data viewer because it has such a simple interface.
When you open up QField, you have one, two, three buttons, basically. So really a lot of map and very few user interface. As of data processing, we just introduced in the latest version that came out a couple of weeks ago, we introduced the possibility of actually running a workflow. Now, I think this will help people understand what’s possible and what the process might look like. So I’ve got an example here. Let’s imagine for a second that I’m a GIS specialist sitting in an organization, maybe a council, and I have people that need to go out and collect data. So I’ve got access to all my data sources through QGIS.
And let’s pretend for a second I’m using a server. I’ve got a few flat files, maybe shape files, maybe a geo package here and there. And I’m using OpenStreetMap as a base map. What’s my process? How do I give this to people to go out and collect data based on these layers that I’ve got here? So the most comfortable way to go that way is in your QGIS, you install the QFieldSync plugin, which will then take care of pushing your project via cloud solution to the people that you want to have access to this project. The people in the field will open the project on their devices, will go out, will report whatever you ask them to report, will push a button to synchronize back the data, or you want to.
And the data come back through QFieldCloud. They will be pushed directly to your Postgres database that you have in the packet. From the office where you are with QGIS, you get access to your Postgres database, you see your data, and you have everything automatically back from the field. Wow. It sounds like this should be a paid service. Is it? Because this idea of data, I mean, QGIS is a free piece of software.
I’m sure QField is also because it’s based on that. What about this QFieldCloud? Is that paid or is this a free service? Both of them. It’s both of them. So QFieldCloud is also an open source tool. We open source all the code under MIT license.
So you could just take your own QFieldCloud, run it on your own server, and there is no cost involved. You’re just doing whatever you want with it. Or we are offering software as a service, which has a community account as well, which is free, which is limited in storage, but basically the complete functionality is there. Or if you want to increase storage or have multiple permissions within teams, and you want to set up an organization then it’s a paid service. Yeah. And just to be clear, I am pleased that it is a paid service. I am pleased that people like you make money and pump it back into open source software and have a job and can create an ecosystem around this.
So I have nothing against paid service. I’m just trying to clarify for myself and for the listeners, okay, which bits are paid, which bits are free. So I appreciate that. That is actually the best way to support QField development is to get a QFieldCloud account because that’s how we are kind of moving money from QFieldCloud into development in QField itself. Because when we want a new functionality, either we have somebody that comes up and says, I’d like this functionality and I’d like to sponsor it. Or we have this big advantage of having QFieldCloud that gives us the possibility to take the income from the service and reinvest it to make QField even more advanced and more, you know, with more features that are being asked. We have an idea platform where people can share what they would like to have in the new versions.
And these are the kind of things that we look at and decide, okay, there are many people asking for, I don’t know, the possibility to directly digitizing something without having to open up an attribute table. We are looking into this and we might want to finance that, or we’d like to have background tracking for position. These are kind of things that came from ideas of users and that we either try to get people together in a sort of almost a crowdsourcing-like, crowdfunding-like, or we just reuse funds that we get through QFieldCloud to implement those things. I think there’s probably a lot more to say around this topic and maybe we should wait for later on the conversation. I’ve got in my notes here, it’s like, why open source? And then I’ve got slash being a good open source citizen. I think that might lead to a lot of the topics that you’ve just sort of touched on there.
But I want to stay with this idea of the workflow, because I think it’s important for people to understand, like, well, what can I do with this? If I don’t have QFieldCloud, what can I do? As I said before, that would be the most comfortable way. If you don’t have QFieldCloud, the QFieldSync plugin, you can also use it for a cable export. So it will just create a folder where you have your project in it. It will kind of consolidate everything that you need within there. And then you can just take the folder, zip it up and send it by mail to whoever you want.
The person can just open it up and it will be usable directly in QField. You can put the QGIS project that you have on some sort of, I don’t know, Google Drive, whatever other drives there are, like Dropbox or whatever. Yes, those sort of things. And people will be able to import the project into QField and just open it up. You can even just send data to someone. So imagine you mentioned shapefiles. I prefer mentioning geopackages.
But let’s say for the sake of things, you’re sending somebody a shapefile via email and this person opens it. QField will be smart enough to do it automatically. An OpenStreetMap base map behind it and actually render your shapefile as a viewer. Or not only as a viewer, you can then even edit it. There are many workflows that you can have. If you’re not using QField Cloud as a service or your own instance, you will just have to take care somehow to get the data back unless you are having QField directly connecting to a database. If you’re doing that, then you get the data automatically back.
Obviously, it will not work as soon as people are offline. Yeah, yeah. That all makes a lot of sense. I just want to go into a little more details around packaging a QGIS project. So I mentioned flat files and a PostGIS server that I was using and OpenStreetMap as a base map. Will it cache all these, presumably as a geopackage? Is that of the project?
Is that how it’s all packaged up? And I’m wondering about caching of a base map as well. When you are in QGIS, you can say on every layer, what action do you want to happen when it gets packaged? So for example, you can say that the layer should be made or your PostGIS layer should be made offline available, which means it will drop it into a geopackage for the area of interest that you want. And you will not share the connection to the PostGIS layer. Or you could say to keep connection to the database itself, and then it will not do anything. It will just leave the project as is and have the connectivity to the database directly.
Regarding base maps, we are not doing an immense amount of automation there. There is a way to basically create offline tiles in QFieldSync itself. But QGIS has already a lot of tooling for doing that. So usually we just rely on the people doing it that way. Yeah, okay. We talked a little bit before about connecting other sensors. So yeah, it sounds amazing that I can go out to the field and I can update a table, for example, or I can edit in the map itself.
So actually that this line is over here, instead of over there, this polygon should look like this. What about other sensors? And when I think about other sensors that are in my camera, my mobile device itself, so we can take images, we can record audio. But I’m also thinking about things like external GPS and perhaps other sensors, maybe like laser range finders, for example. Yeah. So short answer is yes to everything you just said. A bit more explanation is maybe required.
So all your internal sensors, camera, video, audio are supported directly. External GNSS or GPS, as people kind of know it, are also supported in various methods, depending with external position receivers, it’s always a bit tricky depending on what the receiver itself is delivering. But if it delivers standardized NMEA strings, which are whatever they should be delivering, then it works. If they are delivering strange things, then usually there is an app in between from the manufacturer or the device that then converts it to positioning. So you can use most devices. We do actually have a page on the QGIS. org website where we do have some brands that are going, that went through certification as certified hardware.
So that’s a partner program we have. They want really to have their devices tested for every release, and we are doing that so that we are sure that those devices are working correctly. And then the third part you mentioned, the laser range finder, for example, that’s what I mentioned before, that we built in QGIS to get it back in QField. We had a user that needed to get his radioactivity ground levels measured live while moving around the field, adding those values to the points that were being digitized at a regular interval. So the person was moving around and capturing a point every five seconds and creating a map basically of radioactivity of the soil where they were moving on. And so we built a system where you can read any serial string that is coming in, you can configure QField to understand it and to map it directly into an attribute. So very, very powerful kind of things.
There is on the website of QGIS. org, you see a video and a documentation of the person actually moving around, and then they configure the rendering of the map to automatically show intensity related to the other values around. So you get this heat map of radioactivity changing while the person is moving because it gets into more radioactive areas than others. So pretty powerful kind of things, yes. Wow, I can imagine if you could hook this up to things like, if you were measuring noise, for example, and then biking around a city or, I don’t know, if you had another sensor that was measuring air pollution, that would be really interesting, or temperature as well. Yes, basically anything that delivers serial kind of output, just sort of the standard on those professional devices will be able to be read in. More sensors, I mean, all the Android devices or the mobile phones have a lot of sensors in them as well.
And those, we will look into if to use them or expose them because they’re not doing serial. So there is a lot that can be done already. Yeah, so one of those sensors, at least on iPhones, is the ability to scan. So they have a built-in LiDAR, some of them. Is that something where you can see some potential, or have you already implemented that? Well, the LiDAR within iOS is something we look at. Unfortunately, on the Android counterpart, so if you look at Samsung, they dropped their LiDAR scanners from the flagship phones.
Apparently people were not using it as much as we imagined. So we don’t have a direct support for in the sense that you are going to generate a point cloud and you can kind of immediately render it. But if you open up the dedicated camera or the dedicated app for creating point clouds, you’ll get the dataset that you can then just open up in Qfield itself, or you can attach it as an attachment to a point. So you can say, I was in this point and I did this scan, and the scan is kind of the, like you did pictures, videos, and a scan. Is that true for images and audio and video as well? I’m wondering how that is attached to those points. Is it a blob that is written directly into the database?
No, we’re not doing blobs in the database. We are storing relative paths, basically. So your project will have a home where you have the QGIS file, and there you have the DCIM folder where you get all the picture in there. You have an audio folder where you get all the audios and so on. And then if you are using Qfield Cloud, Qfield Cloud will just take care of slowly in the background, also synchronizing those. So if you do a manual sync and you say just push data, it will just push the geo kind of part of the data directly, and everything else gets pushed in the background. So you don’t have to wait until your high-resolution video has been synchronized to the cloud.
You can keep on working, and it will just do background synchronization of audio and video and pictures. How does the sync process cope if I have multiple users editing or perhaps working and updating within the same sort of geographical area and maybe the same objects? So again, I have to make a distinction. If you take the comfortable way with Qfield Cloud, the cloud will identify conflicts and not commit those and allow you to decide what to do in the cloud itself. So you’ll go to app.qfield.cloud, you log in there, and you will go to your project, and you’ll see that there are conflicts, and it will ask you what you want to do. You then decide to take Daniel’s version or take Marco’s version, and then this one gets pushed to the database or to the geopacket or whatever you originally pushed as a data source. If you’re not using Qfield Cloud, you will receive different geopackets from the people, and you will have to deal with them manually, basically.
So that’s a big advantage of going the cloud way. Every time you talk about Qfield Cloud, you talk about the comfortable way. I’m wondering if you should talk to your marketing person, maybe try and put Qfield comfort in there somewhere. We don’t want to kind of push people towards spending or anything. We’re just telling them that it is a way easier way to work than using a cable to export your data. But we want to be core or hardcore, almost open source, so people can do everything without needing to completely offline if they want to with their project and manage them the way they want. We’re just giving them the chance to work in a much more comfortable way so that their work gets even more efficient.
Because the work on the field itself, so your field workers that are doing digitizing, that’s usually a very expensive time. You’re sending out a team of 20 people doing assessment after a natural catastrophe. That’s a lot of manpower that goes there. If they can just push a button and sync, or even better, you pre-configure the project to just auto-sync, that’s a lot of save time. And the people can do better than just being around and fiddling and sending you an email with their shapefiles, and the other one sends you only three of the six required files that shapefile has. See, there are plenty of places that you can lose work and lose quality just because you want to not use a cloud solution. And that is why we decided to offer Kubeflow Cloud as a service, but also to open source the code so that people could just take it and run it within their department if they want, within their infrastructure.
We can help them roll it out if they want, they can do it by themselves, but we want to really give everybody the possibility to have the best tooling around. And I really appreciate the two options there, and also the fact that you’re not ramming it down my throat by this thing. But I gotta tell you, I feel that pain of people heading out into the field and collecting data and coming back with half of the required files for the shapefile, or a PDF that they’ve scribbled on, or a picture that they’ve taken that they don’t really know where it was. There’s a line there, or something like this, and all these archaic ways of bringing data back to the office. And then it was me, I’ve definitely been in that position myself, where I have to figure out what does this person mean, and how do I sync this back into our other data? Otherwise, it’s basically useless to me. They’ve wasted their time, and they end up wasting my time because I can’t sync the data.
I can’t make it already done. And the risk of every time there’s a, maybe not an extra click, but like an extra step in the process, if this is happening like five, six times a day, 48 times a week kind of thing, it’s also another place where I can make a mistake, and I can end up recording something that’s not correct. And maybe people are going to make incorrect decisions based off that. So I’ve definitely been in that position before that you’re describing. Yeah, we have a very good example. Unfortunately, a couple of weeks ago in Switzerland, there were huge floods, kind of 30 years biggest floods, or maybe 100 year kind of event, where many places were destroyed, much infrastructure was destroyed. And I was in screenshots of Google Maps with an arrow drawn it showing where destruction had occurred.
And we quickly set up a project called the Kewfield Rapid Mapper, open it up to the citizens just to install Kewfield, join a project and start reporting pictures of where things were broken via Kewfield Cloud, which we donated for the project. And then even with publishing behind on our web GIS, where you can see where people have been reporting damage to infrastructure and to property. So technology nowadays can really help a lot in those kind of things and can make a lot of very clean data instead of a lot of data in a messy way. So you mentioned this concept to me in an earlier conversation, and I think it was people are it’s kind of enabling that process. It’s I guess, citizen participation. Yes, I think that was one of the papers that stuck with me from my university reading. There was a, I think it was by a good child, this people as sensor paper was really interesting.
It was back in the early 2000s and kind of resonated with me a lot. And I think that’s why I took a little bit this path of Kewfield, where mobile is very important. I must also say that I love being outside. So I like mobile solutions because I can get out instead of being behind a computer. That’s what I like to, I’d like to push. And that’s a lot of why Kewfield is there where it is now as well. So in terms of use cases, so you mentioned this tool, and you had the situation a few weeks ago in Switzerland.
I remember seeing images on social media about this, especially for one town, which I visited, Samat was flooded a couple of times, I think within a week, it was really devastating. I can imagine people using this in utilities companies, collecting data, bringing it back to the office, updating it, that kind of thing, syncing it. From my experience in the utilities field, that would be a great, I could see a lot of use cases there. I could see things like forestry as well, where you are offline potentially a lot of the time, but need access to your data and don’t want to walk around with a PDF map. But in a previous conversation, you mentioned this, I think it’s worth bringing up a story about the German Indiana Jones.
What is that? Yes. I’m not sure if that’s our internal naming for them. So it’s the German Archaeology Institute, they actually have a project called rapid protection. I don’t know the English name, but basically it’s rapid reaction for when archaeological things are found, and they need protections. Whenever there is an excavation or a building being built or being ripped down and things appear, they quickly spin up this group and it goes there and start taking pictures of all the archaeological finds that are made and have this fantastic way to document digitally everything, where it was found, how did it look like, who found it, and so on. And they do that with Qfield.
You have to imagine, if you look on their website, they have pictures of the last big exercise that they did with 70 people. And you see these people wearing white protection coats and masks and gloves and everything. And then they have a Jonesy, but it’s really, really neat to see what our tool is doing worldwide, how much people are using it. I don’t know if I mentioned before, we have almost a million downloads and are doing 320,000 active users per month. And this number is only on Android. Wow. We don’t know about, obviously, MacOS, Linux, and Windows.
We don’t know because we are just offering download, basically. And iOS, they give some metrics as well. We are already way over 100,000 downloads as well. We launched iOS much, much later. It has been only a year and a half or two, I don’t remember exactly. But it’s a huge user base worldwide that is doing an extremely broad kind of work with Qfield. Are any satellite companies, maybe Earth Observation companies, using this as a way of collecting ground truth data?
I can imagine that being a great use case. Dividing up an area, teams go there and ground truth this for me, so I can use that as input into my AI something. Absolutely. Interestingly enough, that’s a big project that we just started with a company a month ago, where they want to put infrastructure where infrastructure is not there yet. We do for them some remote sensing and machine learning analysis first to identify the places where they then will send people with QFIL to ground truth and see if all the machine learning and remote sensing analysis we did were actually correct. Feedback into the loop, the findings directly via QFIL, and then improve all the machine learning and then send out people hopefully again. So it’s really something that is happening in many different industries, yes.
The reason I mention this, I published an episode a wee while ago called Why You Should Astera. The CEO that I was talking to was saying that the biggest jump in efficiency, I guess, in terms of identifying these leaks, which they do from space, using L-band, was when they closed the loop. So they have an app where people go out and confirm, yes, there is actually a leak. You did find a leaky pipe here, or you did not. He wanted to believe his algorithm could do a better job, could solve this all by itself and continue to see these inaccuracies. But it wasn’t until people went out there, this idea of people as sensors, and confirmed or denied that they got the feedback that they needed to create a better algorithm. And he said the jump was a 40% increase.
I can believe it. Absolutely. It’s such an important part of any industry. It’s to ground truth, to send people out that have know-how. When you are there and you’re seeing the data on the field, or you’re seeing the thing that you care about in the field, with tooling that supports you and with the know-how that usually your field people have, you’re just increasing their productivity immensely. Imagine sending out a technician that knows what he’s doing and then can immediately report back to you what has happened. It’s just a huge increase in productivity. Absolutely. I’d like to move on now.
I think you’ve given us a good overview of what QField and QField Cloud are, how they work, especially through talking about those possible workflows there and some of these use cases. I think that’s been great. I’d like to move on and talk about the future. Are we going to continue to develop this?
Are you finished? What can we expect from this project in the future? Well, I said I started 11 years ago, so it’d be a bit, not sad, but no, it’s not finished. It’s never going to be finished. It’s a continuous improvement. We are continuously looking at what people are doing. We’re getting a requirement by a big national agency right now in Finland.
They are doing all their national cadastre data with QField, so they are bringing a lot of new requirements because they obviously have huge data sets they are doing with QField. Plenty of things that are coming in, plenty of new things that we are looking at to build. I think the greatest news that we have lately for the future is that the last version we released very quietly had a complete plugin architecture in it, which means that anybody can now build plugins for QField and deploy them together with their projects. So imagine, for example, the use case I mentioned before, the Swiss flooding scenario. We quickly built a plugin that added a button to QField to just take a georeference picture because that’s what we wanted from the people to do. We wanted them to be out, push a button, snap a picture, and that was the data we wanted. Could you just explain to me how I’d add that plugin to, not how I’d develop it, but as a user, how do I give that to my surveyors, the people that are going to collect data?
How do they find that plugin? I’ll have to use again the word comfortable. And if you’re going the comfortable way and you are using QField Cloud, basically you put your plugin with the same name as your QGIS. So you put the plugin in the same folder as your QGIS project, you push it to QField Cloud, and it gets deployed automatically to all your teams on the field. Wow. Cool. Yes. So that’s a pretty neat functionality that QField Cloud brings.
We have two types of plugins. We have plugins for projects, which was the one that I just explained. And then there are plugins which are for QField for every project. And those ones you need, there is a install plugin button in QField, and you will need a URL where the plugin is sitting to just copy paste it in, and then it will install it automatically. But I think the other one is really, really neat because you will be able to create workflows exactly as your company or as users need with a specific button that we would have never, ever put in QField because we need to keep QField as generic for everybody as possible. We cannot put a specific user requirement of a user within QField if it’s not useful for everybody. But now with the possibility to write your own plugins, you can do exactly that and deploy it super easily, have it automatically on for all your users in your organization and have them just do what they need to do even more efficiently.
This might be a really naive question, but there’s a lot of plugins for QGIS. Can I take one of them and just expose it to QField? Nope. Plugins in QGIS are written in Python. Plugins in QField are written in the native language of the QField user interface, which is QML and JavaScript. So the disadvantage is yes, you cannot take the plugins that you have for QGIS and use them on QField, but I think that’s a very small disadvantage because the plugins usually have a user interface and that user The big advantage is that there is so much code out there and examples for JavaScript is that you will be really quick in learning how to write those plugins and you are in a completely native environment for QField, meaning that you get all the starting and everything already done for you. Things look nice immediately.
You don’t have to care about that. Yeah. I want to go back to you saying that one of the things you’re really excited about for the future of QField was, you’ve got these big customers. Finland is going to be using it for lots of different things. In terms of open source and being a good citizen, is there any way, does that have a direct impact on, I guess, open source and the has an impact or what is the impact of having big customers? Yeah. Well, both, both. Okay. So I think the question goes one step behind and it’s really what your core values are.
So if you look at the history of OpenGIS, we’ve been committed to QGIS extremely. We are the second company worldwide that does most of the work in QGIS itself. We don’t do anything that is not open source and we are trying to push back as much as we can from our client into QGIS itself. We even set up a sustainability initiative for each client that we have. We kind of take all the hours that they don’t use with us and just put them in bug fixing in QField and QGIS. So I think if when those are your core values, it’s really easy to make an impact through a big client into the open source softwares that you’re using. Because when you believe that open source is good for good and technically, then you just do it naturally.
And that pushes both the satisfaction of your client and the tooling forward. So I think for us is, yes, having big client is a very important thing because it allows us to push a lot of things as well in QField and in QGIS. A very simple example is maintenance work. If you have big contracts, big work that you’re doing, you will easily put in bug fixing or a bit of infrastructure work for QGIS and these and that within those frameworks. When you have very small, very focused contracts develop me that button, it’s really much more difficult to put in those kind of work in the background. I think in 2022, that’s the number I remember, we put 1,200 developers hours from our QGIS sustainability initiative, QGIS itself. That’s almost a full year employee. Wow. Yeah. When you have this kind of core value, then it’s very easy to make an impact.
Firstly, let me say thank you very much for that. Thank you for putting that work back into making things better for all of us, not just any one of us. This is amazing. What I really wanted to help people understand with that extremely clumsy question of mine was that that connection between big customers, your success and everybody else’s success, because it’s within the open source ecosystem. So some people might listen to this and go, who cares if they’ve got a big client? Well, maybe you should care because a big client like this, your success means their success further down the line and continued success and maintenance of tools and the development of this ecosystem. So that’s really what I wanted to help people understand.
So thanks for that. They should absolutely care because if yeah, plugin framework, sensor framework, those kinds of things all happened because we had somebody that was really interested in having those and we put in the effort to get it done on the Qt 6 for QGIS. We also did a lot of work there because somebody that we had as a client was very interested in having that. So there’s, everybody should really care about who else is joining in because the more people are putting into the software itself, the more everybody else is getting out of it. We would never be where we are if QGIS did not have these amazing ecosystem of companies and volunteers around it, putting in all the work that they put in. So not all of us can be a good citizen in the same way you can. Not all of us are developers as an example, but there’s a lot of users out there in the world.
How can they be good citizens in terms of open source geospatial? There is plenty of things that people can do. One of the things that I would say, first of all, is telling people, like helping people by doing, if you’re a power user, you know how to use something, tell people how you do it. Do documentation, help writing the Q field or the QGIS documentation. That’s super helpful. Help translate the documentation. That’s extremely helpful to make people understand that everybody that is helping has a huge impact on the project. Get involved.
There is a website to be redesigned. I don’t know. There is some use case to be rigged. And if you have a good use case, it’s really good to share that as well. So there’s many ways that you can support by doing things. And if not, donate also. There is nothing bad in also giving money to projects in open source. That’s actually something that open source project users are really bad at, is making it clear that there is absolutely nothing wrong in making a project also commercially successful so that it can run itself and it can create new features for its users and so on.
And I think here QGIS is a very, very good example. And Q field also. There’s plenty of things that everybody can do, even if you’re not a developer. Actually, probably much more for non-developers than for developers. Interesting. I would add to that list, being an advocate for the tools that you use and the work that you do. And I think the more people that understand what’s possible with some of these open source tools, the more people we’ve got out there advocating for them, the bigger the ecosystem will become. Absolutely. I think that’s key.
That’s what I try to say by telling and helping. Well, Marco, I want to thank you very much for your time. And I want to thank you for the work that you do. You are literally making things better for everyone. And I really appreciate it. We mentioned these names a couple of times. There’ll be links in the show notes to people if they want to go and check these things out.
Qfield, is it Qfield. org or Qfield. com? Where can they go? Qfield. org. That’s where you get all the information about Qfield itself. Perfect. And what was the name of your overarching company again?
Our company is OpenGIS. ch. Yes. And if somebody wants to reach out to you or connect with you in any way, where could they best do that? With my super difficult to pronounce name, the big advantage is that almost nobody uses it in social networks. So anywhere you look for M. Ernazocchi with CCHI at the end, you’ll most likely find me.
And if not, if not, Marco at OpenGIS. ch is a way easier way to get in. Okay, perfect. I’ll try and find some links and I’ll include them in the show notes so people can reach out if they want. Thanks again for your time. It’s much appreciated. And the work that you do is also incredibly appreciated. Thank you.
Thank you, Daniel.
Apple Podcasts
Spotify
