BostonGIS

Free and Open Source for Geospatial North America (FOSS4GNA) 2025 was running November 3-5th 2025 and I think it was one of the better FOSS4GNAs we've had. I was on the programming and workshop committees and we were worried with the government shutdown that things could go badly since we started getting people withdrawing their talks and workshops very close to curtain time. Despite our attendance being lower than prior years, it felt crowded enough and on the bright side, people weren't fighting for chairs to sit even in the most crowded talks. The FOSS4G 2025 International happened 2 weeks after, in Auckland, New Zealand, and that I heard had a fairly decent turn-out too.

One of the biggest problems open source projects face today is the bus factor problem.

I've been thinking a lot about this lately as how it applies to my PostGIS, pgRouting, and OSGeo System Administration (SAC) teams.

PostGIS Day yearly conference sponsored by Crunchy Data is my favorite conference of the year because it's the only conference I get to pig out on PostGIS content and meet fellow passionate PostGIS users pushing the envelop of what is possible with PostGIS and by extension PostgreSQL. Sure FOSS4G conferences do have a lot of PostGIS content, but that content is never quite so front and center as it is on PostGIS day conferences. The fact it's virtual means I can attend in pajamas and robe and that the videos come out fairly quickly and is always recorded. In fact the PostGIS Day 2024 videos are available now in case you wanted to see what all the fuss is about.

PostGIS Day 2023 videos came out recently. PostGIS Day conference is always my favorite conference of the year because you get to see what people are doing all over the world, and it always has many many new tricks for using PostgreSQL and PostGIS family of extensions you had never thought of. Most importantly it's virtual, which makes it much easier for people to fit in their schedules than an on site conference. We really need more virtual conferences in the PostgreSQL community. Many many thanks to Crunchy Data for putting this together again, in particular to Elizabeth Christensen who did the hard behind the scenes work of corraling all the presenters and stepping in to give a talk herself, and my PostGIS partner in development Paul Ramsey who did the MC'ing probably with very little sleep, but still managed to be very energetic. Check out Elizabeth's summary of the event. Many of her highlights would have been mine too, so I'm going to skip those.

Paul Ramsey and I recently had a Fireside chat with Path to Cituscon. Checkout the Podcast Why People care about PostGIS and Postgres. There were a surprising number of funny moments and very insightful stuff.

It was a great fireside chat but without the fireplace. We covered the birth and progression of PostGIS for the past 20 years and the trajectory with PostgreSQL. We also learned of Paul's plans to revolutionize PostGIS which was new to me. We covered many other side-line topics, like QGIS whose birth was inspired by PostGIS. We covered pgRouting and mobilitydb which are two other PostgreSQL extension projects that extend PostGIS.

We also managed to fall into the Large Language Model conversation of which Paul and I are on different sides of the fence on.

I recently released PostGIS 3.3.3. bundle for Windows which is available on application stackbuilder and OSGeo download site for PostgreSQL 11 - 15. If you are running PostgreSQL 12 or above, you get an additional bonus extension MobilityDB which is an extension that leverages PostGIS geometry and geography types and introduces several more spatial-temporal types and functions specifically targeted for managing objects in motion.

What kind of management, think of getting the average speed a train is moving at a segment in time or collisions in time, without any long SQL code. Just use a function on the trip path, and viola. Think about storing GPS data very compactly in a singe row /column with time and being able to ask very complex questions with very little SQL. True PostGIS can do some of this using geometry with Measure (geometryM) geometry types, but you have to deal with that craziness of converting M back to timestamps, which mobilitydb temporal types automatically encode as true PostgreSQL timestamp types.

Anita Graser, of QGIS and Moving Pandas fame, has written several posts about it such as: Visualizing Trajectories with QGIS and mobilitydb and Detecting close encounters using MobilityDB 1.0.

Elizabeth Chistensen already gave a succinct summary of some of the PostGIS Day 2022 presentations, which you can see here.

There were many case study presentations which involved use of PostGIS, QGIS, OpenStreetMap, and pgRouting (and other extensions that extend PostGIS) as well as many "How to" videos. There were also talks on "How PostGIS is made". I'll highlight some of these, which overlap with Elizabeth's list but different angle of view.

I didn't think last two years Post GIS Day conferences could be topped, but I was wrong. This year's was absolutely fabulous and better than all the others. We had two key conferences going on this year (yesterday). The first was the Cruncy Data PostGIS Day 2022 12 hour marathon of nothing but PostGIS related talks. Many thanks to Elizabeth Christensen and Paul Ramsey for putting it all together and for 12 hrs. The PostGIS day celebrations climaxed with many of us playing capture the flag on Paul's new shiny invention until we managed to crash it.

PostgreSQL 15 came out just last week. To celebrate the arrival of PostgreSQL 15, I will revisit the number one problem people have with PostGIS, how to upgrade your PostGIS enabled cluster, without installing an old PostGIS version.

In a previous article Using pg upgrade to upgrade PostGIS without installing older version I demonstrated a trick for upgrading to a newer PostgreSQL instance from PostGIS 2.2 - 2.whatever without having to install the older version of PostGIS in your new PostgreSQL service. This is a revisit of that article, but with considerations for upgrading from a 2 series to a 3 series.

Fear not. Going from PostGIS 2+ to PostGIS 3+ can still be done without installing the old PostGIS 2+ in your new cluster. Additionally once you are on PostGIS 3.1+, you should never have to do symlink or copy hacks to upgrade say PostGIS 3.1 to PostGIS 3.4. At least not for the same major version.

If per chance you are on PostGIS 1 and trying to jump all the way to PostGIS 3+, then sadly you need to do a pg_dump of your old database/cluster and pg_restore into your new cluster.

This year's PostGIS Day was on November 18, 2021. I celebrated like many others in a PostGIS day virtual conference. You can find the talks and videos at PostGIS Day 2021. Others celebrated with parties, food, and spirits. This was my favorite PostGIS Day conference ever. Each year just gets better.

If I were to sum up this year's conference I would say: A generous helping of code, lots of humor, and lots of people. Thanks Elizabeth Christensen, Paul Ramsey and Crunchy Data for putting this conference together. All talks were really good so hard to isolate just a couple.

The Free and Open Source for GIS conference : FOSS4G 2021 Buenos Aires Online took place September 27th to October 2nd, 2021 online. Many of the videos from the conference (including mine on PostGIS) are at FOSS4G 2021 Playlist viewable online.

One of the new features coming in PostGIS 3.1 is fixed precision support. This new feature will require compilation with not yet released GEOS 3.9. There are a couple of functions already in PostGIS 3.1 that have this new feature -- they are ST_Subdivide, ST_Union, ST_SymDifference, ST_Union, and ST_UnaryUnion as summarized in What's new in PostGIS 3.1. To take advantage of these new to die for features, you'll need to have compiled your PostGIS with development GEOS 3.9 which is planned for release around the same time we get around to releasing PostGIS 3.1. Windows users can download binaries with GEOS 3.9 support from PostGIS windows experimental binaries

The ST_Union feature should improve a lot of cases where people ran into topological exceptions. Perhaps I'll demonstrate that in a separate article once I've given it a test drive.

The PostGIS Development Team released PostGIS 3.0.0alpha1 recently. There are still more changes to be added, like leveraging more of PostgreSQL 12 enhancements. I'll detail the user facing changes you will see in the 3.0.0alpha1 which impact install and upgrade. I hope these changes will make it easier for users to install and upgrade PostGIS extensions.

All these changes were covered in the release email among several others.

In a previous article Using pg upgrade to upgrade PostGIS without installing older version I demonstrated a trick for upgrading to a newer PostgreSQL instance from PostGIS 2.2 - 2.whatever without having to install the older version of PostGIS in your new PostgreSQL service. Unfortunately that trick does not work if coming from PostGIS 2.1 because in PostGIS 2.2 we renamed a c lib function that backed sql functions in 2.1.

Fear not. There is still a way to upgrade from 2.1 to 2.5 without installing an older version of PostGIS in your new PostgreSQL instance. To do so, you need to add a step and that is to remove the functions in 2.1 that are backed by this renamed lib function. In upcoming PostGIS 3.0, we've added this function back and have it throw an error so that even coming from PostGIS 2.1, you can upgrade just the same as you do from later versions.

pgAdmin4 version 3.3 released this week comes with a PostGIS geometry viewer. You will be able to see the graphical output of your query directly in pgAdmin, provided you output a geometry or geography column. If your column is of SRID 4326 (WGS 84 lon/lat), pgAdmin will automatically display against an OpenStreetMap background.

We have Xuri Gong to thank for working on this as a PostGIS/pgAdmin Google Summer of Code (GSOC) project. We'd like to thank Victoria Rautenbach and Frikan Erwee for mentoring.