Next Week: See an Incredible Selection of Cloud-Native Geo Talks
I am very pleased to announce the schedule for the Cloud-Native Geospatial Outreach Event that I’m organizing with the Open Geospatial Consortium next week (April 19th and 20th). We’ll have over 70 5-minute lightning talks, along with 6 in-depth tutorials, from an incredible set of speakers. The event will be entirely virtual and completely free and is spread across time zones to be friendly to a global audience. The talks will also be posted as quickly as possible, so everyone can see the ones that aren’t in the right time zone.
Putting the schedule together was a lot of fun, as there are so many great speakers and organizations represented. We’ve got the three biggest cloud providers (Microsoft, AWS & Google) speaking, as well as the three largest satellite imagery providers (Maxar, Airbus & Planet), and several leading software platforms (Carto, UP42, ESRI & Sentinel Hub). Various governmental organizations will also share what they’re up to, including agencies from the UK, Canada, Philippines, Australia, New Zealand, and Brazil. I’m really looking forward to Gilberto Camara’s talk about Brazil’s INPE and their adoption of cloud-native geo, as he was one of my first inspirations working on open source geospatial software and recently lead the Group on Earth Observations (GEO), so he should bring a great perspective.
We’ll also hear from almost all the stars of the cloud-native geospatial communities, with talks covering a wide range of topics and tools. I’m particularly excited for Emmanuel Mathot’s talk on what Terradue built for the UN Disaster Charter, as it’s probably the most advanced cloud-native geo application that I’ve seen. And one of the other talks I’m most looking forward to is Tom MacWright on how real-time collaboration can work with geospatial data, especially since I’m sure he’s going to show off some of the cool advances he’s made in that realm with Placemark.
Multi-user collaboration online is a great example of how cloud-centric architectures can enable new types of workflows that just wouldn’t have been possible previously. I think there is a tidal wave of geospatial innovation about to be unleashed, and I’m excited that this event will highlight much of it.
Don’t forget to register to be able to see all the great talks. We decided on a single track of talks and tutorials for the entire conference. Most of the talk sessions will contain 6 to 8 talks, the vast majority of which are 5-minute lightning talks, leaving time for questions at the end. We’ve all been at enough virtual events in the last couple of years that we know it’s not realistic to expect everyone to give their full attention for hours at a time, so there are breaks and the deep-dive tutorials interspersed throughout the event.
See the schedule page to explore the full agenda, and read on for more of the highlights.
One of our main goals for this event is to truly have it be an ‘outreach’ event, where we go beyond the existing communities and try to really welcome new people in. I’ll open the conference at 14:00 UTC on Tuesday, April 19, 2022 with an overview of the whole Cloud-Native Geospatial landscape and its potential to make a real difference in the world. And then throughout the conference, we’ll have ~20-minute overview talks from key leaders of each of the major cloud-native geo standards: Matt Hanson will talk about SpatioTemporal Asset Catalogs (STAC), Sanket Verma on Zarr, Howard Butler on Cloud-Optimized Point Clouds (COPC), and Jed Sundwall on Cloud-Optimized GeoTIFF’s (COG).
These are all designed to explain not just the standards themselves but the use cases they solve and the growing ecosystems of tools around them. These will serve as the kick-off to hour-long sessions, that will often center around related talks.
The next talks I’d like to highlight are those from our amazing sponsors. This event wouldn’t be possible without them.
Javier de la Torre from Carto is going to give a talk on the spatial capabilities of Cloud Data Warehouses like BigQuery, Snowflake & Redshift. Javier turned me on to the potential of these to absolutely transform the geospatial landscape, so I’m really excited to hear him share his take with a much bigger audience.
Our other ‘Sustaining Partner’ is Planet, and they’re showing up in force. Sara Safavi, who leads the Developers Relations & Open Initiatives team at Planet, will give an overview of Planet’s contributions to Cloud-Native Geospatial and where Planet is investing next. And a number of Planeteers have talks throughout the event, including Mansi Shah talking about how she leveraged COG’s for an awesome art project, Tim Schaub on the new WebGL capabilities of OpenLayers, and Jennifer Reiber Kyle sharing how to do crop type differentiation using temporal analysis of COG’s.
Development Seed is representing with Vincent Sarago, the ‘COG Tzar’, giving both a lightning talk on the amazing TiTiler project, as well as the first tutorial. I’m also really excited that Aimee Barciauskas will be co-hosting a Cloud-Native Geo Q&A with me (more on that below), and I’m sure you’ll see a number of Dev Seeders helping out with moderating the sessions.
We’ve also got some exciting new sponsors to announce. Amazon Web Services just came in as a sponsor, and they’ll be giving three different talks throughout the program. Mike Jeffe will cover how their Earth on AWS has great public data using COG, STAC & Zarr, Ethan Fahy’s talk goes deep into the CMIP6 dataset (stored in Zarr), and Hunter Carlisle will provide a survey of Geospatial Deep Learning in the Cloud. Kartoza also is a new sponsor, and Samweli Mwakisambwe will be giving a talk on their awesome new QGIS plugin for STAC. I’m also excited to share that Safe Software and Natural Resources Canada are new sponsors as well, and I’m hoping we’ll get some talks from them as well.
Rounding out the talks from sponsors includes Eugene Cheipesh from Azavea talking about Spatial Indexing in the new GeoParquet format, Mykola Kozyr from UP42 sharing two talks about his awesome STAC and COG Tools, and Nathan Zimmerman from Azavea giving an overview of the Python ecosystem for STAC. Plus two more talks from UP42, with Valeriia Zakhlieniuk sharing the value of STAC for the industry and Serhii Hulko talking about the advantages of defining a more strict subset of STAC for increased interoperability.
One of the things I’m personally most excited about is hearing from a number of different organizations on why and how they are adopting these new technologies. For everyone who has built the standards, it’s just awesome to hear about the difference they are making in the world. But even more important it’s good for others who are considering embarking on a similar journey to hear the successes and challenges of similar organizations that have been earlier adopters. I often think about Geoffrey Moore’s ‘Crossing the Chasm’, where he explains how ‘mainstream adopters’ base their decisions mostly on who else has had success with a new technology. I think we’re firmly in the ‘early adopter’ phase of cloud-native geospatial, so we need to build up and share more of these success stories to bring this technology.
I’m excited we’ll have 3 full hours of lightning talks on diverse organizations who have had success adopting one or more of the cloud-native geo standards. And the breadth and impact of these organizations are really impressive. We’ll hear from governmental perspectives like INPE in Brazil, LINZ in New Zealand, Australian Antarctic Division, Digital Earth Africa, Centre for Environmental Data Analysis (CEDA) in the UK, NOAA in the US, the Philippine Space Agency (PhilSA), and the Meteorological Service of Canada. With Google, Microsoft, and AWS we have representation from all major cloud service providers. And satellite data providers including Maxar, Satellite Vu, GEOSAT, Carbon Mapper, and Planet will share why they’re all embracing cloud-native geospatial. Interesting organizations like Anglo American, the Global Wind Atlas, Rexy.ai, Sinergise, Weatherlayers, and UP42 round out the sessions dedicated to organizational perspectives, though you’ll also see a number of other interesting organizations scattered throughout the program.
We’ve also got a great set of tutorials for those who want to dive deeper into one of the topics of Cloud-Native Geospatial. These will be an hour of hands-on time with an expert, exploring a particular tool or diving deeper on a standard.
- COG’s: Vincent Sarago kicks off the workshops with ‘Good COG/Bad COG’, which should be a great introduction to how to recognize, create and use COG’s.
- STAC API search with CQL2: One of the newest capabilities of STAC API is the use of the powerful CQL 2 filter language. Phil Varner from Element 84, who has been leading the final push of STAC API to 1.0.0 will take participants through its capabilities.
- Zarr: Get a great introduction to this workshop on Zarr and its capabilities (more details coming soon)
- Pangeo Forge: Much of the innovation around Zarr is from the Pangeo community platform, and this will be a deep dive into one of the key tools, showing users how to transform NetCDF data into cloud-optimized analysis-ready data as Zarr.
- Microsoft Planetary Computer: Looking for a platform that does COG, STAC, Zarr, and COPC in a globally scalable, fully hosted online platform that includes a ton of open data? Search no more: Tom Augspurger will show you how to get started with it.
- stac-fastapi: One of the leading STAC API implementations is also one of the most customizable. Join this tutorial to learn how to create your own customized STAC API using this great python project.
- STAC, COG, and Coiled: In a tutorial called ‘How to turn satellite data to insights at scale’ Basile Goussard from NetCarbon will share how they are using the Pangeo stack of tools to realize insights like mean land surface temperature.
The other session that’s in the same vein as tutorials is the Cloud-Native Geo Q&A session I’m hosting with Aimee Barciauskas. We’re not going to prepare any material, as we just want to give a place for people to ask questions about any cloud-native geo topics. Our hope is that this is a very welcoming space, where any n00b question can be asked. I imagine that some of the various community experts will drop in and help with some of the questions as well. I have no idea how this will go, but if it goes well then we could try to repeat it the next day.
Rounding out this post are a few other sessions and talks I’d like to highlight. One topic I’m most excited about is ‘cloud-native vector’, a topic I dug into recently, and that is seeing lots of progress. There is going to be one session centered around the topic, with Javier from Carto kicking it off, and then a number of talks on GeoParquet, where there’s been a lot of recent activity. But be sure to catch Björn Harrtell’s talk on FlatGeobuf, as it is the more mature ‘cloud-native vector’ format which is definitely worth highlighting. Cloud-native tiling formats will also be discussed, with a talk focused on COMTiles, though it will also compare a number of options including PMTiles, FlatGeobuf, Tapalcatl, Cotar, COMTiles. And then Brandon Liu, who created PMTiles, will be giving a really interesting talk covering a bunch of practical learnings he gained from designing PMTiles, but that are applicable to any format following the cloud-native geo philosophy.
I’m also excited that ESRI is joining the conference, with two talks. One will dive deep into I3S, which is a web and cloud-optimized format for large 3D data. And the other will highlight the use of STAC and COG in their imagery products. The final talk I’d like to highlight is the one on ‘Image Hub at Anglo American’, by Robin Wilson. It’s been cool to see him join the community recently and have some great success applying an open stack of technology to solve some real-world problems. So it should be a great case-study of STAC, COGs, TiTiler & Dask.
If any or all of this sounds interesting don’t forget to register! You’ll then be sent the login information when the event starts. And a big thanks to all the speakers for taking the time to put together and deliver great talks.