Welcoming New Collaborators to the Cloud Native Geospatial Ecosystem

Details on the Sprint & Outreach Day agenda and awards

Chris Holmes
Aug 31, 2020 · 7 min read

I wanted to share some background and more details about the Cloud Native Geospatial Outreach Day and Sprint (1 week away! September 8th — signup here), which has grown out of the SpatioTemporal Asset Catalog (STAC) community sprints. One of our goals for STAC has been to make it a truly collaborative community, and one that is welcoming as possible. We believe that the best standards are forged from diverse use cases and perspectives coming together and collaborating. We started with a gathering of 25 people from 14 organizations and have always sought to bring more people into the fold.

I think we’ve done well at bringing in diverse technology perspectives, but it’s bothered me that the majority of people who join our sprints all look the same. I’m not sure that we’re worse at inclusion than the general geospatial tech industry, but it is a goal of mine to do better in the communities I help lead and to play an active role in welcoming more truly diverse contributions to the broader geospatial & tech industries. I believe geospatial tech has the potential to help us tackle some of the biggest challenges that face us globally, and my experience has taught me we will not live up to that potential if the technology behind it is built from a narrow perspective.

So we are going to try out some experiments with this sprint, to try to do better. I’ll admit I’m a bit nervous about it not working out quite right, or getting something wrong, but I suppose that is the nature of an experiment, and that trying with a willingness to fail is the only way to get better.

One of the traditional ways we’ve used successfully to bring new people into our community is by using any extra sponsorship money to fund ‘travel grants’ to join our in-person sprints. These sprints have been the driving energy behind STAC. Every single new peak of activity on the spec occurred at a sprint:

And each brought in more contributors. But travel grants make little sense this time, as there’s no possibility of an in-person sprint during a global pandemic. So this is our latest thinking.

More beginner-friendly sessions

In each of the last couple of sprints, we’ve added more session to help people new to the STAC community get up to speed. These started as just an informal review of the spec with a long Q&A session that let people ask any question in a very friendly environment where ‘there are no bad questions’. But they have always presumed at least some background knowledge of geospatial and software. This time we are planning for some of our ‘Introductory Sessions’ to not require any background, introducing topics like Machine Learning on Imagery, and Remote Sensing, with lots of support for beginners. I wish I’d done more to organize this ahead of time to give people more time to prepare, but I’m really thankful for the great presenters who have signed up so far (and please let me know if you are interested in sharing your knowledge with new people, submit here or get in touch). Our plan is to record the content and make it available to all, and to iterate and improve in future sprints.

Enabling easier first-timer contributions

A big part of participating in open communities is actually contributing in a way that helps improve things. With STAC the bar to contribute has been pretty high, generally requiring deep geospatial and software development knowledge, to be able to edit the spec itself or to make software. But there is a myriad of ways that people can help the whole STAC and Cloud Native Geospatial Ecosystem, so for this sprint we’re working to try to make it much easier for new people to contribute in a meaningful way.

The first is that we are setting up Azavea’s Groundwork to enable anyone to help with data labeling of satellite imagery to create Machine Learning Labeled Training Data. The output is a STAC catalog, so everyone annotating clouds in the tool will be making meaningful contributions to the ecosystem. We’ll be organizing it into a ‘competition’, with awards for top contributors, and we’ll have more details about it soon, in a separate post.

Beyond that, we are aiming to write up all the ways that people can help transform geospatial data to be ‘Cloud Native’ so that it is easier to jump in and help out. Expanding beyond just STAC enables people to focus on Cloud Optimized GeoTIFF conversion, which has a more mature tool ecosystem. We’ll need help converting, upgrading, and creating STAC records, validating catalogs, testing with various tools, documenting where the data lives, creating tutorials, updating the website, and more. Our hope is that people will be able to use some of the skills they acquire in the Introductory Sessions to help out with further contributions.

Community Awards

The final experiment we’re doing in this sprint is to award a number of prizes to help encourage people to get involved and contribute, with some special categories to help welcome new contributions. This came about as usually we use any excess money for travel grants to bring new people into our community, but with a completely virtual event, we have no travel costs. This time we’ve had by far our largest response from sponsors and our core costs are much less. So we’ve been thinking about how to best use that money to further the Cloud Native Geospatial ecosystem, especially by bringing in more data, tools, and especially people into the fold. You can find the full details at on the Cloud Native Geospatial Award Details doc, but the overview is as follows:

Forward Funding Awards for Most Promising Contributions — We will have two $5000 awards that recognize the top projects moving the ecosystem forward during the sprint. One will be selected by the participants of the ‘software sprint’, and the other by a small panel of judges from the top sponsors. The goal will be for the money awarded to enable the project to continue on past the sprint, so the contributors can justify spending additional time on it.

‘Newcomer Grants’ — We’ve been brainstorming on how to best replace our traditional ‘travel grant’ with something that works during the pandemic. We’ve decided to award four ‘newcomer grants’ that include $1000 to help them continue their work in the community, as well as a full travel grant to the next in-person sprint. One is focused on new contributors from African countries, sponsored by Digital Earth Africa and the World Bank, and another two on newcomer women, one sponsored by Arturo and the other by SparkGeo & Planet.

Community Recognition — The next set of awards aims to recognize an array of contributions. We aim to have a nice custom STAC jacket (like Patagonia or Arcteryx) for top recognition. You can see the full list of awards; some of the highlights include:

We’ll also be making STAC-branded hooded sweatshirts for an additional 30 or so contributors, including everyone who delivers an ‘Intro Session’. And finally, anyone who makes a meaningful contribution to the Cloud Native Geospatial ecosystem will be awarded a t-shirt (with a soft cap of 200 to keep us within budget). See the award page for more information. And we do understand that not everyone values swag in the same way, so awardees will be able to select a gift card as an alternative ($200 for jackets, $50 for hoodies, $20 for t-shirts).

Join us

All of this is to say that we want you, yes you, to join us, even if you know nothing about STAC or Cloud Native Geospatial. We think there are some cool things happening, that have real potential to positively affect the world, and we want to share. The main day for new people is September 8th, and we hope you’ll join us to contribute during the sprint that follows — we promise to help anyone interested in contributing! Sign up at https://forms.gle/3Kx7cffMZov52MH88, and please share this with anyone you think might be interested. We can’t guarantee that everything in the sprint will be a smooth introduction to our world, but we’ll do our best to meet you more than halfway. And you will hopefully help teach us how to do this better in the future.

Radiant Earth Insights

Earth Imagery for Impact