PostGIS 2.0 Released

This is major news: after 26 months of development, the open source geospatial database PostGIS 2.0 has been released. PostGIS is one of the best spatial database system there is. It even has been recently identified as better than Oracle Spatial.

The new features:

  • Raster data and raster/vector analysis in the database
  • Topological models to handle objects with shared boundaries
  • PostgreSQL typmod integration, for an automagical geometry_columns table
  • 3D and 4D indexing
  • Index-based high performance nearest-neighbour searching
  • Many more vector functions including
    • ST_Split
    • ST_Node
    • ST_MakeValid
    • ST_OffsetCurve
    • ST_ConcaveHull
    • ST_AsX3D
    • ST_GeomFromGeoJSON
    • ST_3DDistance
  • Integration with the PostgreSQL 9.1 extension system
  • Improved commandline shapefile loader/dumper
  • Multi-file import support in the shapefile GUI
  • Multi-table export support in the shapefile GUI
  • A geo-coder optimized for free US Census TIGER (2010) data

Open Source Routing Machine: High-Performance Routing Engine

Andrew made me aware of the Open Source Routing Machine (OSRM), a lightning fast routing engine built on OpenStreetMap data with draggable routes.

From their main page: "The Open Source Routing Machine (OSRM) is a C++ implementation of a high-performance routing engine for shortest paths in road networks. It combines sophisticated routing algorithms with the open and free road network data of the OpenStreetMap (OSM) project. Shortest path computation on a continental sized network can take up to several seconds if it is done without a so-called speedup-technique. OSRM is able to compute and output a shortest path between any origin and destination within a few miliseconds. Since it is designed with OpenStreetMap compatibility in mind, OSM data files can be easily imported. A demo installation (currently offline) is provided by our friends at Geofabrik. OSRM is under active development.

The key features of OSRM are:

  • High Performance Routing Algorithm
  • Easy import of OSM data files.
  • Written entirely in C++ and available under the GNU Affero General Public License for anyone to use.
  • Ability to handle continental sized networks.
  • Influenced by current and ongoing academic research. [more info]"

It's certainly not the first time we talk about routing with OpenStreetMap data. For example, in 2008 we talked about the OpenRouteService. And you'll get much more by heading to the routing page in the OSM wiki.

PROJ 4.8.0 Released

The OSGeo announced the release of PROJ 4.8.0. PROJ is, like GDAL/OGR, an open source component at the core of many geospatial software. The last major release was in September 2009.

The major items according to the announcement:

  • Added the Natural Earth projection.
  • Added HEALPIX, rHEALPIX and Icosahedral Snyder Equal Area projections.
  • nad2bin now produces "CTable2" format grid shift files by default which are platform independent.
  • nad2nad removed, use cs2cs for datum shift operations.
  • projects.h no longer installed as a public include file. Please try to only use proj_api.h.
  • Add pj_get_spheroid_defn() accessor.
  • Added an alternate version of pj_init() that takes a projCtx (execution context) structure to address multithreading issues with error management and to provide a support for application hookable error reporting and logging.
  • Upgrade to EPSG 7.9. Some changes in ideal datum selection.
  • JNI bindings reworked, org.proj4.Projections deprecated in favor of org.proj4.PJ.
  • Added preliminary vertical datum support.
  • Fix various multithreading issues, particular in datum grid handling code.
  • Added support for the +axis= option for alternate axis orientations as part of a coordinate system (used for TM South Orientated support).
  • +proj=omerc implementatioin replaced with code from libproj4. +rot_conv flag no longer works, and some coordinate systems (ie. Malaysian) will need to use +gamma instead. "epsg" init file updated accordingly.

Here's the full changelog.


PostGIS vs Oracle Spatial: PostGIS Wins

Unless you're new here, you've heard of PostGIS several times. It's really hard to find documentation about spatial database benchmarks since, if I'm not mistaken, proprietary software licenses often prohibit publicly sharing such benchmarks (am I right?). And I guess there's at least one good reason for this: doing fair benchmarking is difficult to accomplish. But here's one (currently available) benchmarking report comparing PostGIS (PostgreSQL) and Oracle Spatial.

From the conclusions of the 46-pages report: "From the experimental results that we saw, we can conclude that Postgres performs better than Oracle 11g both in the Cold Phase and Warm Phase. Though in few queries Oracle 11g performed better but on the whole Postgres overpowered Oracle 11g. In the warm phase in 3 out of 4 queries Postgres performed significantly well, from this we can conclude that Postgres has better automatic memory management capabilities and page replacement policies. [...] Since Postgres uses the underlying GEOS (Geometry Engine - Open Source) library functions for implementing the geometric operations whereas Oracle 11g implements them on its own, and since in majority Postgres performs well, we can conclude that GEOS geometric algorithms are more efficiently designed than Oracle 11g. And also Postgres planner is more efficiently designed to take advantage of any available indexes to use in queries for achieving better performance whereas in Oracle 11g we saw that we have to specify them explicitly through functions."

Here's what Paul Ramsey of the PostGIS fame has to say about it: "Methodologically there are two obvious issues: one is that the Oracle database was on Windows while the PostGIS database was on Linux; the other is that neither database got any tuning, they were both installed and run with default parameters. However, this is one of the nicer comparisons I have read: concise, focussed and with enough technical detail to evaluate what's going on."

Ten Things You Need to Know about Open Source Geospatial Software

All Points Blog offers an entry named Ten Things You Need to Know about Open Source Geospatial Software.

Here's the list, and head to the entry to read further explanations:

  1. Open source geospatial software refers to GIS, GPS, spatial data management and related developer tools and end user applications delivered with an open source license.
  2. An open source license must meet the definition developed by the Open Source Initiative.
  3. Open source software is written by a community rather than a development team associated with a single software company. [...] Some do this as part of their “day jobs,” while others volunteer.
  4. The Opposite of Open Source is Closed Source or Proprietary Software, not Commercial Software
  5. Open Source Software is “Just” Software
  6. OSGeo is the Body for Open Source Geospatial Software
  7. Open Source and Open Standards-based Software are not the Same Thing
  8. Implementing Open Source Software May Cost Money
  9. Software developers and software users mix and match open source and proprietary software all the time. Esri’s ArcGIS, for example, includes the open source GDAL (raster handling) library. 
  10. Open source licenses are designed for software, not data. There are other licenses appropriate for data.

TinyOWS 1.0 Released: Open Source High Performance WFS-T Server

Via Daniel I learned about the 1.0 release TinyOWS, an open source high performance WFS-T server.

From their website: "WFS service and therefore GML output streams are becoming widely available in quite common SDI architectures. The transactional WFS profile (WFS-T) provides ability to give a writable access on each feature stored on a given WFS server, through a standardized Web interface, in an OGC compliant way. Exchanged data volumes can easily become huge when data is retrieved, thus a high performance WFS-T service is now a real need to provide a good Quality of Service in web-based GIS applications."

OpenJUMP GIS 1.5.1 is shipping - with reliable CORE and valued PLUS edition

The past week OpenJUMP team has released OpenJUMP 1.5.1. As usual we added functionality and fixed bugs - since our last release. However, we made a hopefully user friendly decision: we will distribute a CORE and a PLUS version.

The CORE version is the usual OpenJUMP version. The PLUS version will contain OpenJUMP plus a number of useful plugins. For instance Sextante for raster analysis (still Version 0.6), CadPlans PrintingPlugin for… well.. printing,  DXF driver, PostGIS connector (for writing to PostGIS), Graph Analysis and Topology tools - that can be used for Quality Assurance, the Chart Plugin for creating charts, Beanshell editor etc.

Both versions come as handy zip (all platforms), install exe (Win) or jar-based installer (for Mac and Linux).

For the MacOSX users, we finally fixed a long pending bug, and saving layers is now possible again, without cumbersome procedures.

to download directly check here:
if you just want to know more:
(and don't forget our wiki… for documentation etc.)

Please note that some of the additional plugins are not yet translated and that we renamed some menu functions and moved some into other menus. Save as SVG is   now only available in PLUS as it required some special other functions. Currently we are working on documentation so you can more easily find them again.

some more highlights since OpenJUMP 1.4.1:
- new Fill-Polygon-Tool, an editing tool to create a polygon from a closed area
- new Noder Plugin : nodes/splits lines and polygons
- lots of new icons
- new Buffer Plugin including all options of JTS API
- more Attribute Calculator buttons/functions
- new Geometry Conversion tool
- added new options to the BlendLineStrings tool
- new AdvancedMeasurePlugin with more capabilities
- when closing OpenJUMP and non-saved in-memory layers exist a dialog pops up and data can be saved in shapefile or jml format
- SQL queries: saving, refreshing, editing, and max features
- a new Union/Dissolve/Merge plugin unifying old Union and Dissolve
- reworked and improved Auto-Assign PlugIn (takes now negative numbers and decimals)
- … more can be found in the changes.txt file

Canadian 2011 Census Data Freely Available

Earlier this week, Statistics Canada started to make freely available data from the 2011 Canadian Census. Here are the release dates for the reminder of the data.

​While this is excellent news, I invite you to read the informative article: "Upon playing with the data dissemination interface today, my concerns were re-affirmed.  The data are free but not necessarily accessible, in the sense that the methods used to disseminate and discover these is complicated, unclear and there are some favourite geographies missing – most notably Dissemination Areas (DA) [editor's note: it's there, see at bottom of the full article] while others are hidden – Census Tracts (CTs). For example, if you go to the Census Profile and you want to look up 5 cities at once you cannot! You can only look up one city at a time, which also means you can only download one geography at a time.  There are over 2000 cities in Canada and if you want to know who the top 30 are in terms of population, then its “Houston we have a problem!” sorta."

GeoNode 1.1 Released

We first mentioned it two years ago, and now the open source GeoNode 1.1 has been released.

What GeoNode is? "GeoNode is an open source platform that facilitates the creation, sharing, and collaborative use of geospatial data. The project aims to surpass existing spatial data infrastructure solutions by integrating robust social and cartographic tools. At its core, the GeoNode has a stack based on GeoServer, Django, and GeoExt that provides a platform for sophisticated web browser spatial visualization and analysis. Atop this stack, the project has built a map composer and viewer, tools for analysis, and reporting tools."

And what version 1.1 has for us: "

  • Improved documentation
  • Support for GeoServer 2.1, including:
    • GeoWebCache integration
    • direct Shapefile-to-PostGIS import from the GeoNode upload form (thanks to Matt Bertrand)
    • speed improvements to the way GeoNode manages GeoServer configuration
  • Support for GeoNetwork 2.6
  • Various UI improvements, including:
    • a new user profile page listing the user’s layers and maps
    • a “Get feature info” tool to identify feature attributes when viewing maps
    • improved map transitions and performance
  • Increased coverage in GeoNode’s automated test suite
  • Better feedback from admin tools (thanks to Ariel Núñez)
  • Installer for Ubuntu (thanks to Ariel Núñez)
  • Numerous bug fixes thanks to support from partners at last year’s roadmapping summit
  • Translations in…"

SpatiaLite 3.0 Released with Excel Spreadsheets Support

I failed to find much information about it, but SpatiaLite, the geospatial version of SQLite, reached version 3.0 about a month ago. Anyone knows where to find release notes? I find also funny that on SpatiaLite homepage, it is clearly stated that spatial is not special! :-) Yes, I'll share a followup to my previous entry on the topic (thanks for your feedback!).

On the SpatiaLite topic, here's a blog entry named Spatialite and Excel on talking terms: "The recent stable version of Spatialite, 3.0, supports linking to and importing Excel spreadsheet tables. Read on to see how it’s done. The developers of spatialite have added a driver for *.xls files (thru the FreeXL library ). You can either link to, or import a single sheet from an Excel file [...]"

Syndicate content