The history of QGIS: from a PostGIS viewer for Linux to one of the world’s most important open-source geospatial platforms

There is an especially interesting story inside AEC software.

A developer simply wanted to be able to view PostGIS data from Linux.

So he started writing a small application.

He was not trying to build a global competitor to ArcGIS.

He was not creating a startup.

There was no huge corporate roadmap.

That small project would become QGIS.

Today QGIS can:

• view data;
• edit;
• analyze;
• automate;
• produce professional cartography;
• work with Python;
• process rasters;
• consume web services;
• handle spatial databases;
• visualize in 3D.

And all of it remains open source.

Before QGIS: professional GIS was expensive

During the 1990s and the early 2000s, professional GIS was strongly associated with proprietary software.

There were excellent tools.

But the cost could be an enormous barrier for:

• students;
• universities;
• small governments;
• consultancies;
• researchers;
• independent users.

Meanwhile something important was happening inside open source.

Technologies such as these were appearing and maturing:

• Linux;
• PostgreSQL;
• PostGIS;
• GRASS GIS;
• GDAL.

There were very powerful pieces.

But a modern, accessible desktop experience that could join them was missing.

Gary Sherman

The central figure in the birth of QGIS was Gary Sherman.

In 2002, Sherman wanted to view data stored in PostGIS from Linux.

The options available did not feel simple enough for that goal.

So he started programming.

2002: Quantum GIS is born

Official documentation places the birth of the project in May 2002.

In June it was established as a project on SourceForge.

The original name was Quantum GIS.

And its initial goal was surprisingly modest:

provide a GIS viewer for PostGIS data.

From viewer to GIS

Very soon the project began to grow.

Because once you can see geographic information, the next need appears:

view
↓
edit
↓
analyze
↓
create maps
↓
automate

QGIS began adding support for:

• vector;
• raster;
• shapefiles;
• GeoTIFF;
• spatial databases;
• web services.

Qt and C++

QGIS was built mainly using:

C++
+
Qt

Qt made it possible to develop a cross-platform graphical interface.

That helped QGIS run on:

• Linux;
• Windows;
• macOS.

This point was crucial.

An open-source GIS project that only worked well on Linux would have had more limited professional adoption.

QGIS ended up being truly cross-platform.

GDAL/OGR

A fundamental piece of the ecosystem is GDAL.

GDAL can work with an enormous number of geospatial formats.

Thanks to ecosystems such as GDAL/OGR, QGIS can open and transform an extraordinary variety of data.

Conceptually:

QGIS
↓
GDAL / providers
↓
Shapefile
GeoTIFF
GeoPackage
GeoJSON
databases
...

The power of QGIS does not come only from the code that lives inside QGIS.

It also comes from the open-source geospatial ecosystem around it.

GRASS GIS

Another important historical integration was with GRASS GIS.

GRASS had existed long before.

It was an extremely powerful geospatial platform.

QGIS made it possible to use many of its capabilities from a more accessible interface.

This pattern would repeat:

QGIS as an integration point for geospatial engines.

The community starts to grow

QGIS did not have a single company controlling all development.

Contributors included:

• developers;
• universities;
• consultancies;
• governments;
• companies;
• users.

This community structure produced a model very different from traditional proprietary software.

2007: Python changes QGIS

A huge milestone arrived in 2007.

QGIS 0.9 introduced Python bindings.

This opened development to a much larger audience.

Before:

C++

After:

C++
+
Python

And that allowed any developer with Python knowledge to start building:

• scripts;
• plugins;
• tools;
• automations.

PyQGIS

The Python API is normally known as PyQGIS.

With PyQGIS you can control programmatically:

• layers;
• geometries;
• selections;
• processing;
• symbology;
• layouts;
• projects.

Conceptually:

QGIS
   ↓
PyQGIS
   ↓
automation

This turned QGIS into more than an application.

It also became a geospatial development platform.

OSGeo

In 2007 QGIS entered the incubation process of the Open Source Geospatial Foundation — OSGeo.

In March 2008 it graduated formally.

Being part of OSGeo strengthened governance and its integration inside a broader open-source geospatial community.

2009: QGIS 1.0

After almost seven years of development came QGIS 1.0 “Kore”.

The project timeline records version 1.0 at the end of 2008, with the stable stage consolidating in 2009.

The version provided something fundamental: a stable API.

That gave more confidence to developers building solutions on QGIS.

Plugins

One of the features that would define QGIS was its ecosystem of plugins.

A plugin can add capabilities for:

• geocoding;
• hydrology;
• cadastre;
• imagery;
• networks;
• BIM;
• databases;
• processing.

This produces an effect similar to the AutoCAD ecosystem:

platform
+
plugins
=
thousands of specialized workflows

But with a fundamental difference:

QGIS is open source.

Processing Framework

Another important leap was the integration of Processing.

Processing can run algorithms from multiple providers through a common system.

For example:

QGIS
├── native algorithms
├── GDAL
├── GRASS
├── SAGA
└── plugins

The user can connect algorithms in:

• graphical models;
• scripts;
• batch processing.

This turns QGIS into a kind of orchestrator of geospatial engines.

2013: QGIS 2.0 Dufour

On 9 September 2013 came QGIS 2.0 “Dufour”.

This version mattered for many reasons.

But one of the more curious ones was the name.

Until then the project was officially Quantum GIS.

With QGIS 2.0 the project formally dropped the word Quantum.

From then on it would simply be QGIS.

Official documentation even clarifies that “Quantum” never had a particular meaning.

QGIS 2 becomes professional

The 2 series consolidated capabilities such as:

• cartography;
• labeling;
• Processing;
• Python plugins;
• databases;
• raster;
• analysis models.

For many organizations QGIS stopped being:

a curious free alternative

and became:

a real professional tool.

2018: QGIS 3.0 Girona

On 23 February 2018, QGIS 3.0 “Girona” was released.

It was an enormous renewal.

The project described the change as a major review and cleanup of the platform.

QGIS 3 modernized a large part of:

• APIs;
• UI;
• architecture;
• Python;
• Qt.

The 3 generation would last eight years and become the base of an enormous global expansion.

3D

Over time QGIS began to stop being only a 2D map.

It incorporated 3D visualization.

This is especially important for AEC because it can represent:

• terrain;
• DEM;
• buildings;
• point clouds;
• meshes.

GIS and BIM start to come closer.

Point Clouds

Point clouds have become fundamental in:

• surveying;
• LiDAR;
• infrastructure;
• inspection.

Modern QGIS has tools to view and process this kind of information.

That brings GIS closer to reality-capture workflows.

What is QGIS used for in AEC?

1. Hydrology

It can be used for:

• DEM;
• watersheds;
• slopes;
• drainage networks;
• precipitation rasters.

And to prepare data for:

• HEC-HMS;
• HEC-RAS;
• SWMM.

2. Roads

QGIS can manage:

• preliminary alignments;
• cadastre;
• right-of-way;
• terrain;
• constraints;
• utilities.

It does not replace Civil 3D.

It plays another role: geospatial context.

3. Cadastre

It is widely used for:

• parcels;
• properties;
• attributes;
• spatial queries.

4. Infrastructure networks

It can represent:

• drinking water;
• sewers;
• energy;
• telecommunications.

When combined with spatial databases a much more powerful system appears.

5. Hazards

It can be used for:

• floods;
• landslides;
• earthquakes;
• vulnerability;
• exposure.

6. Environmental

It can work with:

• land cover;
• ecosystems;
• watersheds;
• protected areas;
• pollution.

PostGIS: a return to the origin

There is something elegant in the story.

QGIS was born because Gary Sherman wanted to view PostGIS.

Today:

QGIS
+
PostGIS

remains one of the most powerful open-source geospatial stacks.

PostGIS can store geometry inside PostgreSQL.

Then multiple users can work on:

database

instead of exchanging individual files.

QGIS Server

QGIS is not limited to the desktop either.

There is QGIS Server.

It can publish projects and data through geospatial web standards.

For example:

• WMS;
• WFS;
• other OGC services.

This makes it possible to build complete GIS infrastructures around the same ecosystem.

QField and mobility

The ecosystem also includes mobile tools such as QField.

This makes it possible to take geographic information into the field.

A workflow can be:

QGIS
↓
project
↓
mobile
↓
field survey
↓
synchronization
↓
QGIS

Very useful for:

• inspections;
• surveying;
• inventories;
• assets.

IFC and BIM

QGIS is not a BIM authoring tool like Revit.

But the GIS and BIM worlds are converging.

It is increasingly common to combine:

IFC
+
terrain
+
parcels
+
networks
+
city

The value of QGIS is placing the asset inside its territorial context.

QGIS and Python

For AEC, PyQGIS can automate complete workflows.

For example:

read DEM
↓
clip
↓
reproject
↓
generate raster
↓
export for a hydraulic engine

Hours of manual work can become a reproducible script.

QGIS and an AEC harness

QGIS is practically perfect as the geospatial component of a harness.

An agent could:

search layers
↓
reproject
↓
intersect parcels
↓
process DEM
↓
create map
↓
export GeoJSON
↓
feed another engine

And do it through:

• Python;
• Processing;
• commands;
• services.

This allows QGIS to be more interesting than a GUI: an automatable geospatial engine.

2026: QGIS 4

On 9 March 2026 came QGIS 4.0 “Norrköping”.

It was the first version of the 4.x generation.

One of the most important internal changes was the migration of the core toward Qt6.

The project noted that the new base would improve:

• maintainability;
• performance;
• security;
• access to modern libraries.

QGIS 4.2

In July 2026 came QGIS 4.2.

And currently, as of September 2026, the published version is QGIS 4.2.2 “Belém do Pará”, released on 28 August 2026.

The available LTR version remains QGIS 3.44.14 “Solothurn”.

This shows something interesting.

A program that started as a PostGIS viewer in 2002 reached its fourth major generation more than twenty years later.

QGIS vs ArcGIS

The comparison is inevitable.

ArcGIS represents an enormous commercial ecosystem.

QGIS represents an open, community ecosystem.

It is not correct to reduce the discussion to:

one is free
one is paid

The difference includes:

• development model;
• licensing;
• APIs;
• ecosystems;
• cloud services;
• commercial support;
• governance.

In many workflows they can even coexist.

Open source does not mean no companies

QGIS is open source.

But around it there is an economy of:

• consultancies;
• developers;
• support;
• training;
• plugins;
• implementation.

That shows that:

open software and business are not incompatible ideas.

The real strength of QGIS

It is probably not a single function.

It is the combination:

QGIS
+
GDAL
+
PostGIS
+
GRASS
+
Python
+
plugins
+
OSGeo

QGIS works as a human and programmatic interface to an entire geospatial universe.

Timeline in brief

2002
Gary Sherman
PostGIS viewer for Linux
        ↓
Quantum GIS
SourceForge
        ↓
vector / raster
cross-platform
        ↓
GRASS integration
        ↓
2007
Python bindings
OSGeo incubation
        ↓
2008
OSGeo graduation
        ↓
2009
QGIS 1.0
        ↓
plugins
Processing
        ↓
2013
QGIS 2.0 Dufour
"Quantum" disappears
        ↓
2018
QGIS 3.0 Girona
        ↓
3D
point clouds
PyQGIS
ecosystem growth
        ↓
2026
QGIS 4.0
Qt6
        ↓
QGIS 4.2.2

Conclusion

QGIS is probably one of the best examples of how an open-source project can grow.

It was not born trying to dominate world GIS.

It was born because one person needed to view a PostGIS database.

Then developers arrived.

Then users.

Then plugins.

Then Python.

OSGeo.

Processing.

3D.

Point clouds.

Server.

Mobile.

And in 2026, QGIS is already in generation 4.

For AEC its importance is enormous because practically all of our projects exist somewhere in territory.

A structure can be designed in ETABS.

A road in Civil 3D.

A flood in HEC-RAS.

A network in EPANET.

But QGIS can place all of those results inside the same geographic world.

And that ability to connect:

data + territory + automation

is exactly what makes GIS an essential piece of modern AEC.

References

1. QGIS Documentation. Foreword.
2. QGIS Project. Visual Changelog QGIS 1.0.
3. OSGeo. QGIS Annual Report 2007 / Python bindings.
4. OSGeo. QGIS Incubation Status.
5. QGIS Project. QGIS 2.0 Changelog.
6. QGIS Project. QGIS 3.0 Girona.
7. QGIS Project. QGIS 4.0 Norrköping.
8. QGIS Project. Download QGIS.
9. QGIS Project. PyQGIS API.

Evaluation and limitations

Treat the notes below as evaluation context, not as a verified calculation or code check.

Use it as a professional GIS platform: data, territory, plugins, Processing, and Python automation. Open source does not mean no companies and no governance. In many workflows it can coexist with ArcGIS.