In+ersec+ion for Spatial People

Earlier this week the FOSS4G 2011 organizers announced that registration is now open. Being hosted in Denver, Colorado, in September 12-16, 2011, it's expected to be the largest FOSS4G conference ever and clearly be one of the geospatial conferences to put on the agenda.

From the main FOSS4G 2011 website: "Recent years have seen substantial changes in the geospatial industry. One of those changes has been the growth in maturity and adoption of free and open source solutions. In many cases organizations are using a mixture of open and closed source solutions. With the growth in interest in open source solutions, combined with the fact that FOSS4G has not been in North America since 2007, we anticipate a great audience for FOSS4G 2011, around 1000 people."

The call for speakers was released last week. The workshops and initial tutorials were also announced. Of note, OpenStreetMap's State of the Map conference will take place immediately before the FOSS4G Conference.

Slashgeo has been a media partner to the FOSS4G conference in 2009 and 2010. We're in talks to again provide coverage on-site this year. You'll find our previous coverage with a search.

I told you it would eventually be published, there it is. My personal notes of my attendance at the FOSS4G 2010 conference. They have been edited to remove content specific to my organization. Please find other FOSS4G 2010 entries here.

• Opening session
  • Opening remarks by Lorenzo Becchi
    • 364 abstracts submitted
    • 869 (final number) attendees, the largest FOSS4G conference ever
    • 61 countries represented
    • Canadian representation in severe decline
    • ESRI and Google are sponsors, of course in addition to AutoDesk. So we have quite a few major proprietary software providers involved.
  • Miguel Montesinos' talk: Beyond open source
    • FOSS4G vs what? Miguel argues it's "vs privative" software
    • Business cases discussed
  • Schuyler Erle's talk: How crowdsourcing changed disaster relief
    • Software is useless without data
    • Commons-based peer production, a longer name for "crowdsourcing"
    • Haiti disaster and OpenStreetMap, a well known success
    • Neogeography is fast... which is why it beats paleogeography/traditional GIS for some contexts
  • Arnulf Christl's talk: The state of OSGeo
    • eGov means using electrons to help govern ourselves
    • Fun disconnected talk

• Vertical datums - Frank Warmerdam
  • Vertical datums: meteorological data is intrinsically 3D
  • Review of GIS datums basics
  • Tidal datums, local, means, etc
  • Orthometric vertical datums height is measured from the Geoid... ngvd29 navd88 igld85 are national and regional implementations
  • Geiod, Equipotential gravity surface, varies by up to 100m from geocentric ellipsoid.. Global EGM96
  • Nothing do with directly with ground level by itself
  • No obvious pivot vertical datum in EPSG or anywhere else
  • OGC WKT code for vertical-cs representation
  • No good OGC support of vertical datums, no compound datums which would include vertical datum
  • Frank is the main developer of GDAL/OGR, of which we use operationally at CMC
  • Geoid vs ellipsoid heights: most people don't know what they have and use
  • Liblas vertical datum support
  • Planning to add GDAL and Proj.4 support for verticals datums
  • Some software support vertical datums in CSMAP from Autodesk and in Geotoolkit
  • Ellipsoid datums are ok for most GIS applications

• Interactive METEOSAT education platform for meteoreological applications with FOSS4G - Vasile Craciunescu
  • ESA eduspace
  • New site (November 2010) using OpenLayers + Extjs
  • User uploads of observations using Extjs
  • Batch image processing with GDAL, similar to what we do
  • They can't support Internet Explorer because of it's lack of svg support
  • They support weather symbols in color, eg. wind barbs
  • Provide weather observations info boxes in Extjs directly on the web map
  • www.asrc.ro/imeteosat_beta/
  • Not planning to support WMS at the moment, only flat jpegs

• OpenLayers: SOS and INSPIRE - Bart van den Eijinden
  • Met Bart at FOSS4G 2009
  • Developed an SOS client into OpenLayers
  • Weather data into SOS standard displayed on an OpenLayers web map
  • Included in OpenLayers 2.9, needs proj4js
  • No convenience layer for the SOS protocol in OpenLayers
  • The example is on the OpenLayers examples dev site under SOS client
  • Limitation of SOS: one request per offering (obs value)
  • INSPIRE is similar to Canada TBS mandatory requirements: WMS, CSW, and WFS are in use (mandatory?)
  • OpenLayers now can flip axis order to accommodate WMS 1.3 support
  • For Netherlands geozet project, they had to support Internet Explorer 6 and 7, it only serves points at the moment
  • Using clusters in OpenLayers
  • New openlayers.strategy.filter capabilities.. Coming in OpenLayers 2.10

• OSGeo General Annual Meeting
  • Tyler Mitchell, executive director of the OSGeo
  • State of the OSGeo, board panel
  • OSGeo is adding value to the software projects
  • Budget, projects
  • From 442 to 657 code contributors, 301 contributors in the last 12 months, for a total of 13 millions lines of code
  • Peer-reviewing committee ?
  • Officially 40 chapters, much more in formation
  • Next FOSS4G is in Denver 2011
  • The future of the OSGeo? Community vs organization, how do we measure our success? By the adoption of FOSS4G software?
  • Google Summer of Code: Google indirectly supporting OSGeo again, some are great successes
  • The OSGeo Board presentation
  • Funding challenges...
  • Board is 90% America+Europe and 100% men at the moment

• Michael Gould of ESRI: Shades of grey and opportunity for collaboration
  • That's ESRI at FOSS4G opening session, nothing less
  • Reminder, ESRI uses GDAL/OGR
  • Future, as seen by ESRI, is a blend of proprietary and open source software
  • ESRI codeplex, 14 open source projects at the moment, including an ArcGIS Editor for OpenStreetMap
  • Open standards and specifications, open platform support (Python, Linux, GDAL), but no open source core of course
  • ESRI non-profit program recently announced, to make their (proprietary) software more accessible
  • Geoservices REST specification , white paper published on September 7
  • Extending the reach of FOSS4G with ArcGIS compatibility
  • ESRI claims being committed to the FOSS4G community, not black and white, shades of grey

• Another game of chess? - Ivan Sanchez
  • Topics: games, hacking and military GIS
  • Game theory: same topic as O'Reilly's presenter at EC's GeoSymposium 2010
  • Rather interesting and geeky, but little relation to geospatial
  • More money vs more software
  • Ivan claims that the GPLv3 is forcing people into better options, because going from free to close is not optimal
  • Happiness leads to better software and better performances, not money
  • For GIS, the only winning move is to share

• OSGeo research and education - Helena Mitasova
  • GIS education dominated by proprietary software, need more flexibility and portability
  • Not getting rid of proprietary software, but expand
  • At NCSU FOSS4G-based courses are mandatory
  • WinGRASS was instrumental to adoption, because machines were Windows to accommodate ArcGIS
  • Both using GRASS GIS and ArcGIS for assignments
  • Arizona State University: GRASS GIS, multispec, imagej, and more...
  • Polimi, Milan, Italy: very strong FOSS4G use
  • In research, FOSS4G doing good, since allows required flexibility for research projects.. Showed quite a few examples
  • FOSS4G stimulates innovation and new developments
  • Google SOC program is really helpful to students

• Progress on the implementation of INSPIRE - Ioannis Kanellopoulos
  • Ioannis working for the European Commission
  • Keywords are metadata, interoperability, network services, data and services sharing, coordination
  • Implementation rules vs technical guidance: domain is evolving so fast - binding vs non-binding (mandatory elements)
  • Plenty of stuff to become operational in 2011
  • INSPIRE augments/extends some of the OGC standards, because they require more

• pgRouting - Anton Patrushev
  • PostGIS-based
  • Works with various datasets
  • Costs are dynamically calculated, not precalculated
  • Openvrp.com - Open Vehicle Route Planner

• SOS vs WFS - Carsten Hollmann
  • Conclusions is using both WFS and SOS
  • SOS 2.0 will be a major improvement

• The State of PostGIS - Paul Ramsey
  • The main geo-enabled RDBMS, even compared to many commercial ones
  • Paul is always a populate speaker
  • Similar talk to the one given at FOSS4G 2009
  • PostGIS, Oracle Spatial, MS SQL Server 2008, all support the OGC SQL-SF standard. While others such as MySQL don't.
  • Postgresql is enterprise database, including all features required for real enterprise operational databases
  • Managing changes in shapefiles is a pain in the ass... PostGIS enables analysis
  • In 1.5, released this year, GUI loader
  • 2.0 coming in 2011
  • FME was the first proprietary software to support PostGIS, but now many many, including ESRI, Autodesk, PCI Geomatics, MapInfo, and many more
  • One of the big successes of FOSS4G software
  • So much cost effective, especially when you need multiple databases. Which is the case at MSC
  • PostGIS's point in polygon in polygon is very fast
  • Paul could not complete his talk... lack of time

• WMO Information System - Timo Proescholdt
  • WIS "weather information system": catalog of available products and network of data exchange
  • INSPIRE requirements are applied
  • They're using the ISO 19115 metadata standard
  • Poorest talk so far.. little geo content, little FOSS4G content
  • They use their own network, not the Internet for (some) data transfer in their WMO IS
  • Service-oriented architecture
  • Implementation in progress... over 15 years !
  • Using GeoNetwork Open Source
  • What's the status of Canada on contributing to the WMO WIS?
  • Sru2jdbc, open source implementation of Sru
  • Contributed to jzkit
  • Using FOSS4G because it means members can use it without licenses/cost issues
  • OpenWIS
  • WMO using the ISO metadata standard for geospatial data, including meteorological data
  • Only the metadata is on the WMO IS, not the actual data: users are redirected to national portals

• PostGIS meets the third dimension - Olivier Courtin
  • New surface types
  • There are existing standards, OGC and 2 draft ISO standards
  • Poor English from Olivier... but content ok
  • Polyhedral surfaces
  • A bit technical on the implementation
  • It's in the PostGIS trunk
  • Some known implementations of many features and requirements already
  • Discussed 10 main issues, well documented
  • No 3D topology operations existing other than in draft standards ... SQL/MM standard
  • Aimed to be partly included in PostGIS 2.0
  • 3D topology library concerns more than PostGIS, and the responsibility should be shared with other FOSS4G applications

• Drift-X WPS: Pesticide atmospheric dispersion Web-GIS - Nicolas Bozon
  • Using QGIS and WPS from the ZOO kernel
  • Pesticides from wine growers
  • Wind dispersion and deposition
  • Consider local topography in their models
  • Their application is for air quality
  • They're using a Gaussian atmospherical dispersion model
  • Extent of several square km
  • Model coded in Fortran 77...
  • Grid of 33x33 km
  • Dem is SRTM-DEM
  • Drift-x is a Python QGIS plugin
  • Users can remotely launch the model, which requires a few minutes to run (How do they get the wind values?)
  • Fortran support in ZOO 1.0
  • Compliant to the WPS protocol
  • Chaining of scripts, projection, interpolation, format conversion, update map file, etc
  • Are planning to support more complex dispersion models
  • This is essentially a web interface with a web map to launch and display dispersion model results
  • Not using common WPS profiles (if such a thing exists)

• OGC WMTS and OsGeo TMS standards - Joan Maso
  • Lack of interoperability in all existing tile mapping solutions. Including from Google and Microsoft, and Yahoo, and Amazon S3
  • OGC went with WMTS after OSGeo TMS existed
  • OGS WMTS was approved in 2010, started in 2006, long process
  • TMS is square tiles, while WMTS is rectangle, which caching size for different scales
  • TMS is pure RESTful implementation, WMTS is not, but was influenced by the TMS REST capabilities
  • What makes OGC WMTS better? Nothing.. at least that's what the speaker said
  • SOAP is strange for tile map services
  • Interesting slide on OSGeo vs OGC as creators of standards, seems like OSGEo wins: less secrecy, more open, wikis, much faster, less constraints of the other OGC standards
  • There's a MOU between OSGeo and the OGC on the topic, recognizing the value of mutual collaboration

• TimeMapper: using animated SVG in a WMS to visualize moving object data - Barend Kobben
  • Vector animations generated automatically from the data
  • svg is an open W3C standard, supported by all major browsers, even IE9
  • Svg includes smil declarative animation
  • For their prototypes, they used moving icebergs
  • Rimapperwms, which supports svg, provides outputs in svg from WMS compliant requests! Its a java server
  • From temporal data, using the ISO 8601 standard
  • Demo done, using moving points

• WMS Shootout
  • WMS benchmarking, 8 teams this year instead of the usual 2
  • This year with Cardcorp, GeognoSIS, Constellation-SDI, ERDAS APOLLO, GeoServer, Mapnik, MapServer, Oracle MapViewer, QGIS MapServer.
  • Both open source and proprietary solutions compared with the same datasets, the same WMS requests, same output formats, etc.
  • Testing 18 gigs of vector data.
  • Testing 120 gigs of raster data.
  • Using real Spain data.
  • From 1 to 64 parallel clients requests (2152 requests total).
  • Main winners are the users, because we are the ones who will benefit from the improvements done to the WMS servers (one rule of the Shootout is that modifications must become available to the users, I guess this also applies to proprietary software competitors).
  • This year, benchmark includes bottlenecks from CPU, disk access, network access and remote database access.
  • If you wonder, these guys really went deep in the benchmarking analysis (such as OS-level caching), take a look at the slides. They do try to do a serious benchmarking comparison.
  • There was serious differences in servers which were disk-bound vs disk-unbounded (CPU-bound).
  • Tests included dbf files of over 2 gigs.
  • Of course, there is performance in terms of time to answer requests, but there is also the quality of the rendered tiles, which is more subjective.
  • All contestants agree that this was a lot of work to participate to the Shootout, but it was worth.
  • GeoServer got many great improvements for the sake of the Shootout, same for MapServer.
  • Mapnik: even built their own new WMS server named Paleoserver.
  • Oracle MapViewer: found quite a few bugs and bottlenecks, and yes, they plan to participate next year.
  • QGIS MapServer: integrating desktop GIS and server. Missing time to get consistent results.
  • Constellation-SDI: it's open source software (LGPL). First serious work in performance, many challenges found.
  • Cadcorp GeognoSIS: Windows-only server. Added many improvements for the sake of the Shootout.
  • ERDAS APOLLO: exciting benchmarking project. Made many improvements too. They're refusing to publish results because CPU-bound results cannot be compared with Disk-bound results. Yeah, contestants can step aside from displaying results for any reason they see fit.
  • And now the results...
    • MapServer wins first test, Mapnik second
    • second test, MapServer first, GeoServer second
    • third test, Constellation, map server GeoServer all doing good
    • fourth test, MapServer generally faster
    • fifth test, still MapServer ahead
    • on the last test, Mapnik doing good
  • Overall, one can say that MapServer can be considered the fastest of them all at the moment. Of course, considering that some servers were able to cache in RAM better than others.
  • For next year, the whole WMS process will start much earlier.
  • The show ended by the WMS-song.. pretty funny.
  • Obviously, the full results will be available soon on the OSGeo website.

• GeoServer WPS - Andrea Aime
  • Alper Dincer talk canceled
  • In WPS, processes are defined by the WPS developer
  • WPS is very flexible
  • Andrea is always a good speaker, this matters in talk selection
  • GeoServer debuted in 2008
  • Processes sources include JTS, Sextante, GeoTools, GRASS
  • There's a WPS request builder
  • Why WPS in GeoServer while there is the ZOO Project? Because of WPS integration with GeoServer
  • WPS GeoServer can make requests to remote servers
  • Can do the crunching locally, do not send back the results and only access it via local WFS or WMS: pretty wise and can significantly enhance performance when all results aren't needed in XML
  • WPS can fix many shortcomings of WFS, such as generalization, bounding boxes, data statistics, etc
  • Scripting WPS with geoscript.org
  • No asynchronous execution yet
  • Via the GUI, XML WPS request are built automatically
  • Demo time
  • Part of GeoServer beta
  • GeoServer interface is really user-friendly
  • Wow, another great talk by Andrea

• TileCache, GeowebCache and MapProxy comparison - Till Adams
  • Static vs dynamic caches
  • WMS-C vs WTMS
  • Some supports caching other standards than WMS, including WFS and kml
  • Provided a comparison table
  • This is a major issue of WMS serving performance

• Bringing the gap between open source tools and proprietary data sources - Dean Hintz
  • Talk from Safe Software, the guys of FME, a popular canadian ETL tool
  • Talk is mostly a big FME demo, linking with various data sources, including PostGIS, GeoServer
  • Obviously, they claim ites best to mix open source software with proprietary software... their software! There are a few open source ETL tools, such as GeoKettle and Spatial Data Integrator, but probably not as mature as FME
  • Not really a convincing presentation, but reality is FME is a popular product
  • Talk also INSPIRE-oriented
  • Streaming CAD data in Google Earth from GeoServer with FME on-the-fly processing, leaveing the original data untouched

• Authentification and authorization of OGC services with GeoShield - Milan Antonovic
  • Geoprisma is another solution
  • Learned about GeoShield at FOSS4G 2009
  • Room literally overflowing, despite speaker being not comfortable
  • Server-side is pretty complex, using java servlet, GeoTools, java persistence lib and Postgresql
  • There is a real nice web admin GUI
  • OGC services are encoded
  • Includes WMS and WFS security
  • Long demo done
  • Planning to add SOS and WPS protocols
  • GeoShield is not available to the public yet, but should be in the coming months
  • Done by SUPSI
  • Running inside Tomcat
  • Comment from the audience: filtering is illegal in terms of the WMS spec, next WMS standard version will include a tag to let know users that there will be a filter applied

• Permission-based access to OGC web services with 52North - Jan Drewnak
  • Discussed with Julien-Samuel of the GeoPrisma developer team, GeoPrisma being more client-oriented
  • Integrates with other 52North protocols, such as WSS and PDP, policy enforcement service and policy decision service
  • Works for WPS, WMS, WFS, SOS
  • There are filters on users, roles, IP addresses, etc
  • Integrated license-based access, so users can access data after license agreement and paying fees if applicable, no human administrator required... Interesting but doesn't really apply to what we do at EC
  • Presentation by 52North, which is a competing organization to the OGC
  • This is a more dynamic access control

• SpatiaLite, the shapefile of the future? - Pirmin Kalberer
  • SpatiaLite is spatial extension to SQLite
  • Single file storage
  • Spatial data types and functions
  • Launched in March 2008
  • File based, so no server process
  • Well proven and widely used
  • Now with SQLite 3.7 reading and writing can proceed concurrently, which was not the case not so long ago - this is not activated by default, for backward compatibility issues
  • It supports OGC simple feature types, SQL-SF
  • Indexes for spatial data
  • Supports networks and routing
  • Raster data also supported
  • Based on the GEOS lib
  • Spatial queries and operations
  • Supports OGC metadata
  • Built-in shapefile import and export capabilities
  • Coordinate reprojection via the proj.4 lib
  • There's a spatiality GUI and GIS GUI, two GUIs for SpatiaLite, which are real nice too
  • Supported by GDAL, QQGIS, GeoTools, FDO, etc
  • Many advantages over shapefiles
  • This could be an interesting solution instead of messing with shapefiles at CMC
  • Long interesting demo done
  • SpatiaLite databases can be edited directly in QGIS for instance
  • Performance can be less than shapefiles if we're talking raw access, such as dynamically serving to MapServer
  • SQLite exists for Android, but not SpatiaLite yet
  • There is no available benchmarking of the ESRI File GeoDatabase vs SpatiaLite done so far, that would be interesting!
  • There is no API for the ESRI File GeoDatabase at the moment, it has been promised "soon" for the last three years
  • Pirmin works for Sourcepole
  • Frank Warmerdam comment: questioning the stability of the SpatiaLite, Spatialite is not there yet: Alessandro, the SpatiaLite maintainer, has not opened the development of SpatiaLite as much as it could be in regards to community consensus and backward compatibility

• Squeeze: MapServer and KML - Michael Ross
  • GeoBC (Canada) working with DM Solutions
  • Squeeze tries to address some limitations of Google Earth, such as extent zooming, etc
  • Just a few complex polygons makes Google Earth slow, thousands of of simple polygons
  • Using MapServer 6.0, to be released
  • Much work to make complex vector layers adequately show up in Google Earth
  • Michael claiming some features are better implemented in GeoServer than in MapServer for kml outputs
  • Interesting slide on configuring kml outputs in MapServer 6.0
  • openmaps.gov.bc.ca
  • Tilted WMS and Vertical WMS is in the talks for the next WMS standard version.
  • Tilting WMS layers can cause displaying issues

• Introducing OpenScales - Sébastien Deleuze
  • Nothing more pertinent... pretty rare
  • Open source mapping framework using flash
  • Relies on Adobe for the player, no 64bits OS support, no Linux Flash builder, speaker mentioned that no iPad/iPhone support is not ideal
  • OpenScales at version 1.2 at the moment
  • Modular framework
  • Supports WMS, WFS, kml, OSM
  • Supports 75000 points or 2500 lines with a good user experience
  • API similar to OpenLayers' API
  • Uses Adobe AIR
  • Supports multitouch on mobile devices
  • Openscales 2 in the works, targeted for 2011
  • Prototyping 2.5D support
  • Expecting Adobe to add 3D support to the Flash player later this year

• OpenLayers' future - Tim Schaub
  • Not a lot of competitors to OpenLayers
  • Started as a reverse-engineering of Google Maps
  • In 2006 the context was different... Javascript support was significantly less
  • Version 2.0 came out in August 2006
  • OpenLayers 2.10 released this September
  • Pace of development slightly slowed, but it's quite sustained
  • OpenLayers supports an incredible amount of APIs including proprietary formats and sources
  • Talking about OpenLayers 3
  • Goal one is making it smaller and easy to slim down. That will be a major improvement
  • Future is provide adapters to other js libraries, such as jquery
  • Goal number two is making it faster
  • Third goal is making the library cleaner
  • Fourth goal is making it better
  • Better calls and renaming of functions
  • Live demo done
  • Support for coordinates calls in specific projections
  • Can have much more compact code to use (application)
  • Removing code ambiguity
  • Higher level application events
  • Easier vector styling
  • CSS parser coming
  • OpenLayers wants to support mobile devices natively... OL shown working fine in the iPhone simulator

• Performance considerations in OpenLayers based web mapping - Andreas Hocevar
  • 5 tricks to make open layers faster
  • One is loading OpenLayers the right way
  • Benchmarking done
  • Minified with build profiles... include only the .js that you really need, not the whole 900k, for WMS, it goes down to 150k
  • Two is know your config options
  • Two invisible rows and columns of extra tiles requested for dragging purposes. You can remover this default behavior
  • You can also set it to one tile for the whole extent instead of multiple tiles... this has pros and cons, less requests, but difficult caching... Custom tile sizes
  • Third tip is group your layers, server side or client side... This configuration option speeds up dragging, can be grouped client side with GeoExt
  • Fourth tip, use server side cache... Nothing surprising there, Andreas is using WMS-C caching
  • Fifth tip, combine OGC services wisely
  • Vector layer and selectfeature control
  • Andreas mentioned that Internet Explorer is not modern and not suited for web mapping
  • All source code provided
  • Will salvage this presentation once public

• Distributed spatial indexing in the cloud - Schuyler Erle et al.
  • There is Jeff's MapServer project status report talk at the same time, but dared select Schuyler with his friend, including Paul Ramsey... I couldn't resist listening to these FOSS4G movie stars once again
  • Open talk with people from OpenGeo, SimpleGeo, the Cassandra and GeoCouchDB databases
  • Done without slides... Plenary type of talk with 5 experts in front of us
  • Clustering on the cloud
  • Acid compliance
  • RDBMS require more technical expertise than the NoSQL databases ... but there's a tradeoff
  • "Charting" is getting the database distributed on multiple servers
    • For geospatial databases, there is often manual charting because it's hard to figure out which geodata is close to other geodata... traditionnal techniques don't work for spatial data
    • There no charting in Cassandra, and the performances are horrible if you do complex queries... so its a matter of balance
  • The Challenge for RDBMS is distributing, and for Cassandra, is implement spatial functions efficiently
  • Explaining some index schemes for spatial data
  • Geohash brings 2 coordinates into a single serialized value, so you can easier build some indexes that works one 1D only, such as b-tree
  • Postgresql is already working on distributing the database... so should not be a problem in the future for PostGIS
  • But a a lot of people run queries on huge databases... That's one problem for distributed systems
  • For the users: find what database type is best for your kind of data (SQL vs NoSQL)
  • Eventually, a GeoCassandra database will be released
  • I could not understand every technicality discussed, but enough...

• GeoKettle ETL tool - Thierry Badard
  • Known the guy and the project for some time already, but never actually took the time to learn about GeoKettle
  • In BI (business intelligence), data comes in a variety of formats, generally not geo-enabled, that's the context of their ETL tool
  • ETL tools are clearly an important part of geodata processing workflows
  • GeoKettle is metadata driven
  • First GeoKettle release was May 2008
  • Geokettle 2.0 coming soon
  • They claim already quite a few corporate users
  • GeoETL from Inova is in fact GeoKettle
  • Running parallel vs sequentially: that's the distinction of transformations vs jobs
  • They have a GUI for creating transformations and jobs
  • It does run on Linux and MacOS X, along with Windows
  • Handles quite a few things, including from JTS, topological predicates
  • Kettle is designed to run deployed in clusters
  • Run spoon.sh after unzipping and you're... fairly straightforward installation process
  • Live demo done, using MSC weather observations data!
  • GeoKettle sounds pretty interesting... I'd like to know how it compares to SAFE's FME
  • There's no support for the third dimension, this is upcoming
  • GeoKettle is Java-based

• Closing session
  • OGC's Raj Sing - The tao of information sharing: standards, open source and the future information ecosystem
    • Tao = true nature of the world
    • Europe passed America in terms of OGC membership
    • Geospatial information and supporting technology
    • Software is useless without data (as Schuyler Erle noted in the opening session)
    • Open source and standards is not really about money, it's more about mindshare
    • Geospatial went from "query and analysis" to "view and display" and then to "collaboration"
    • If we optimize for 'collaboration' and we are rewarded by mindshare and that software is useless without data, then... how do we redesign what we do?
  • OSGeo's Tyler Mitchell - Community stats
    • Growth everywhere, mailing lists, local chapters, projects, sponsors, etc
    • 30,000 subscriptions in the mailing lists over 15,000 unique subscribers, while a lot of FOSS4G projects don't host their mailing list on the OSGeo servers and are thus not included in those stats
    • 40 official local chapters, I'm part of the OSGeo-Qc one
    • Trend of software development, 13 million lines of code, 657 committers, with 300 active in the last 12 months... all these numbers are going up
    • FOSS4G attendance is up too... 2008 and 2009 were lower because of the destination (South Africa and Australia) and financial crisis... but still a trend for up and this year is a record with close to 900 attendees
    • Open source adoption is growing... initially from geeks, then to colleagues, to the competition and now government
    • Workshops were sold out
    • Sol Katz Award 2010 given to Helena Mitasova
      • GRASS GIS project involvement along with many other projects... Education, scientific, software, FOSS4G contributions in general
    • Next FOSS4G will take place in Denver... get ready to attend! Judging from the organizing committee... It's going to be a major event.
    • Recap of the WMS shootout benchmarking session
    • There was actually 869 attendees at the event

Here's an attempt to catch up the "recent" geospatial open source news that were published the past three weeks.

Let's start with the 'everything else' category:

• In August there was an introduction entry on QGIS MapServer, a "WMS server for the masses"
• Around the same time, an entry indicating GRASS GIS can make pretty maps
• There was an introduction to MOSSKitt Geo, an UML tool for spatial databases
• Here's an entry on comparing Oracle GeoRaster with PostGIS WKT Raster
• There's news on the OGR MS SQL 2008 Spatial driver
• Shapely 1.2.4 is out and Rtree news
• The OpenGeo Suite Community Edition 2.1.3 was released at the end of last month and the OpenGeo workshops are also available - what a wonderful resource!
• Here's the steps done by ESRI towards open sourcing their Geoportal Extension

Specifically on the FOSS4G 2010 conference, in addition to my own upcoming note, here's some of the many entries on the topic:

While we discussed SpatiaLite in the past (but the archives haven't been imported to this new site yet), here's the notes I took during Pirmin Kalberer's talk at FOSS4G 2010 named "SpatiaLite, the Shapefile of the Future?". In the notes below, Frank Warmerdam is the main developer of the famous GDAL/OGR library. In the discussion, people wondered about the status of the promised ESRI File GeoDatabase API.

• SpatiaLite doesn't replace PostGIS as a real database
• SpatiaLite is spatial extension to SQLite
• Single file storage
• Spatial data types and functions
• Launched in March 2008
• File based, so no server process
• Well proven and widely used
• Now with SQLite 3.7 reading and writing can proceed concurrently, which was not the case not so long ago - this is not activated by default, for backward compatibility issues
• It supports OGC simple feature types, SQL-SF
• Indexes for spatial data
• Supports networks and routing
• Raster data also supported
  

Here are my personal notes for the FOSS4G 2010 closing session. More to come later.

OGC's Raj Sing - The tao of information sharing: standards, open source and the future information ecosystem

• Tao = true nature of the world
• Europe passed America in terms of OGC membership
• Geospatial information and supporting technology
• Software is useless without data (as Schuyler Erle noted in the opening session)
• Open source and standards is not really about money, it's more about mindshare
• Geospatial went from "query and analysis" to "view and display" and then to "collaboration"
• If we optimize for 'collaboration' and we are rewarded by mindshare and that software is useless without data, then... how do we redesign what we do?

OSGeo's Tyler Mitchell - Community stats

• Growth everywhere, mailing lists, local chapters, projects, sponsors, etc
• 30,000 subscriptions in the mailing lists over 15,000 unique subscribers, while a lot of FOSS4G projects don't host their mailing list on the OSGeo servers and are thus not included in those stats
• 40 official local chapters, I'm part of the OSGeo-Qc one
  

First, I want to apologize, I just got Internet access while at FOSS4G, so I couldn't provide good coverage so far... But don't worry, I'll make sure you get all the most important announcements and news, probably some time next week. At least for the elements other Slashgeo editors won't have already shared.

Meanwhile, here's my notes and the results of the famous WMS Shoutout, which just ended minutes ago.

- WMS benchmarking, 8 teams this year instead of the usual 2.
- This year with Cardcorp, GeognoSIS, Constellation-SDI, ERDAS APOLLO, GeoServer, - Mapnik, MapServer, Oracle MapViewer, QGIS MapServer.
- Both open source and proprietary solutions compared with the same datasets, the same WMS requests, same output formats, etc.
- Testing 18 gigs of vector data.
- Testing 120 gigs of raster data.
- Using real Spain data.
- From 1 to 64 parallel clients requests (2152 requests total).
- Main winners are the users, because we are the ones who will benefit from the improvements done to the WMS servers (one rule of the Shootout is that modifications must become available to the users, I guess this also applies to proprietary software competitors).
- This year, benchmark includes bottlenecks from CPU, disk access, network access and remote database access.
- If you wonder, these guys really went deep in the benchmarking analysis (such as OS-level caching), take a look at the slides. They do try to do a serious benchmarking comparison.
- There was serious differences in servers which were disk-bound vs disk-unbounded (CPU-bound).