Archive for the 'Internet' Category

4 Years OpenStreetMap - Details & Interviews


Your Ad Here

Two years ago OpenStreetMap (OSM) was completely unknown in the german-speaking community and central Europe. Today several places in the world have already very detailed and sophisticated maps, produced by volunteers. People uses GPS devices or local knowledge to make existing maps more detailed.

Frustration about the high prices for geospatial data was the motivation for Steve Coast to start the OpenStreetMap project in 2004. Four years later the project developed from an idea to a world-wide map, where thousands of people are participating. Everyone can use the map for free, to embed it on the own web site.

In the past every five months the user doubled, and in the end of 2009 there should be ten times more users than today. And these users will add 200.000 km per day, so the estimations. Indeed, OpenStreetMap is more than a simple streetmap. The map includes information about roads and their environment as well as Points Of Interest (POIs). Questions like, “Where is the next postal office?”, “Where is the next toilette?”, or “Where can I find the next WiFi Access Point?” can be answered. How this can look like shows the POI Control. This web site used OpenStreetMap and analyzes stored POIs.

Of courses some places, especially important towns, are very detailed while other parts are more or less empty. If inhabitants didnt know about OpenStreetMap this can also be seen on the map. For example parts of Spain are not collected. In the USA freely available TIGER data was integrated in OpenStreetMap.

This picture shows a comparison of the city Villach, Austria (pop. 58000) in Google Maps and OSM.

One other important service is routing. An important service which uses OpenStreetMap data is OpenRouteService.org as reported in June.

Following I have summarized important issues form an interview with Frederik Ramm about OSM:

Q: There is Google Maps. Why is there also OpenStreetMaps?

Ramm: Google Maps doesn’t offer their data. You can view their maps, but you can not render maps in your own style. You also cannot put your own routing algorithms on to of the map or calculate the density of letter-boxes for German towns. Google has maps - OpenStreetmap has geodata. That’s an enormous difference. Everyone can participate; a corrected error, a new build road can be found within hours on the map. And finally OpenStreetMap has a free licence, you can use the data for all of your own projects. With Google Maps you cannot do this.

Q: Google has introduced Map Maker and Pedestrian routing recently. Is this a reaction to the OpenStreetMap project, which does not only concentrate on vehicles? Or only a logical improvement?

Ramm: Map Maker a Google services which is based on a technique, that was used for internal data acquisition in India. Google would like to buy map data simply form agencies but for a lot of areas there are simple no useful data available. With Map Maker Google tries to fill these gaps. But we see it relaxed. Google will encourage a lot of people for this project and OpenstreetMap will also benefit. For pedestrian routing it is similar.

Q: Where are the strenghts of the OpenStreetMap project?

Ramm: Free accessibility of data is an enormous potential for creativity. Every few weeks there is somebody who has worked on a software fir OSM. One further strength is that everyone can participate and provide her/his expert knowledge about a local region. The community is the core of the project. Competitors are driving their surveying vehicles. With this technique you cannot get the secret paths between two house blocks.

Q: And where are the weaknesses?

Ramm: There is no complete worldwide coverage. Yet. Further, the community is fixed on map visualisation. This leads to missing Information. There is also no mechanism which can evaluate the completeness and correctness of the data.

Q: What’s about mobile devices?

Ramm: At this years ‘State of the Map’-Konferenz Nick Whitelegg has introduced Freemap Mobile, a J2ME-application for hikers. A complete editor for a mobile platform is not available but moving map and routing works for a lot of applications. Navit is one example which works with the Nokia N810, but also with other Windows mobile devices. Definitively, there is potential.

[via Golem.de]

Written and submitted from Home, using my 802.11g WiFi network.

Semantic Search with Cuil


Your Ad Here

Cuil (pronounced “cool”) is a search engine unveiled on July 28, 2008. Recently several start-ups wanted to establish on the hard search engines market, but also big companies like Microsoft and Yahoo managed it to attack Google. But the semantic search of Cuil tries to make it different. Uniquely, it organizes web pages by content and displays relatively long entries and pictures for each result. It claims to have a larger index than any other search engine, with about 120 billion web pages. Cuil is managed and developed largely by former employees of Google.

Today it started their service and provided relatively good results in my opinion. The search results are not in form of a list, it is more like the layout of an online magazine. Cuil can use 33 million $ Venture-capital. Also Tom Costello is on board, who has worked in the field of search technologies for IBM and made research at the Stanford University.

Written and submitted from Home, using my 802.11g WiFi network.

Could It be that StreetView is comming to Austria?


Your Ad Here

According to GWB a StreetView car was seen in Vienna. Maybe we can soon walk through Vienna virtually. Of course the faces of people and country code plates will be blurred to ensure privacy. But there are still some criticism about this service. Until now this message was not confirmed with a photo. So we will see when StreetView is coming to Austria.

In France Streetview was already launched and StreetView cars are seen in several other European cities including Milan, Berlin, etc.

[via GWB]

Written and submitted from Home, using my 802.11g WiFi network.

Writing KML from my C# Application


Your Ad Here

geKML is a project allowing to generate KML/KMZ documents from a C#.NET application.  geKML consists of .NET classes that will generate a properly formatted KML document that the Google Earth and Google Maps application can understand.  The purpose of these objects are to help the developer to easily construct their desired KML document hierarchy.  These objects will not help the developer decide what hierarchy is best for their application.  Unfortunately you’ll will still need to know how to structure your KML. geKML is hosted on SourceForge and is available for free.

How you can getting started

(original excerpt form the homepage)

The most important class you’ll need to use every time, is the geKML class.  This class will be the one responsible for writing the proper KML header information, and it will also kick off the generation of the rest of the KML objects.

To instantiate the geKML object, you will first need to decide what your “root” KML object will be.  It can be any object that Google Earth will recognize as a root object.  Popular choices are the Feature classes (NetworkLink, Placemark, ScreenOverlay, GroundOverlay, Folder, Document).  For simplicity sake, I recommend starting with the geDocument or geFolder object.

Since most of the Google Earth classes are optional, you’ll have to instantiate most of these objects as you want to use them and then add them to any properties or collections.

Once you have added all of the objects into your geKML.kmlRoot property, you will want to call the geKML.ToKML() method which will return a properly formatted KML byte array, which you can then save to a file or stream from a web server.

Below is a simple example of a basic document (root) and a placemark that exists within that document.  For simplicity sake this example uses hard coded values for properties, but in practice you will probably use values from some dynamic data source.

Welcome to the Google Earth KML API .NET 2.0 Website.  This website will host the official current documentation for the ge-kml project that is hosted on SourceForge.net.  Text versions of the documentation will also be hosted there, but due to scripting and formatting limitations I have chosen to host the official documentation here.  All menu items other than “Home” will take you to the project on SourceForge.

Click on the image to see the class diagramm in full size.

One simple Example from the Homepage:

using System; using System.Collections.Generic; using System.Text; using System.IO; using Google.KML; namespace Examples { public static class Example1{ public static void RunExample(string FileName) { // Use a Document as the root of the KML geDocument doc = new geDocument();  doc.Name = "My Root Document"; //Create a Placemark to put in the document //This placemark is going to be a point //but it could be anything in the Geometry class gePlacemark pm = new gePlacemark(); //Create some coordinates for the point at which //this placemark will sit. (Lat / Lon) geCoordinates coords = new geCoordinates(new geAngle90(37.422067), new geAngle180(-122.084437)); //Create a point with these new coordinates gePoint point = new gePoint(coords); //Assign the point to the Geometry property of the placemark. pm.Geometry = point; //Now lets add some other properties to our placemark pm.Name = "My Placemark"; pm.Snippet = "This is where I put my Placemark"; pm.Description = "I wonder where this is..."; //Finally, add the placemark to the document doc.Features.Add(pm); //Now that we have our document, lets create our KML geKML kml = new geKML(doc); //And our results... File.WriteAllBytes(FileName, kml.ToKML()); } } }

[via: http://gekml.boseefus.com/]

Written and submitted from Home, using my 802.11g WiFi network.

Free Addressdata for Austria


Your Ad Here

An open and downloadable address dataset is useful for many geospatial applications. Therefore the Fachhochschule Nordwestschweiz has developed a platform where users can enter own addresses for benefit of the whole community. Recently FH Kärnten has adopted this idea to create a address dataset for Austria. Until now less than 100 addresses are stored but this could change soon. You can add addresses at the homepage of the portal: http://www.openaddresses.at

Since Google provides a very sophisticated geocoding function including most of the addresses in Austria the success of this project is questionable. One advantage of OpenAdresses is the possibility to download the dataset as CSV or XML. We will see how big the growth of will be.

The Idea:

In the last few years web applications have integrated users more and more in the design, the processing and gathering of information.
Some of the these applicatins are Wikis and famous sites such as youtube, etc.
We had the idea at our institute of launching an experiment to collect a complete set of geocoded address information for Switzerland. This data is supposed to

  • be freely available to all interested companies or private persons and may be used at no license fee
  • be captured and maintained by the Web Community
  • store not only the details of an address but their co-ordinates as well.

Of course there is already a complete geocoded address dataset for the whole of Switzerland - but it is fairly expensive. With the OpenAddress idea we want
to investigate whether it is possible to create such an comprehensive dataset based both on the most recent technologies and the Web Community.
We believe that the more people there are who join in, the less the effort of each individual will be. If many, many people help with their very local know-how, we will be able to achieve this goal.
Besides: it is very easy: just click on the map and enter the details of an address - that’s it. Or use a list of addresses and go sequentially through it.

Of course we would like the individual to collect and enter more than just one address:
This could be: Addresses of the places where you work, you live, meet your friends, have your hair cut or your teeth checked, the most famous restaurant, as well as the most famous restaurant and other popular locations etc.

A few comments:

  • Whoever needs many addresses should collect and enter many. This is especially meant for companies who will use this data in a commercial environment.
  • The project is based on trust: We neither check nor log what is collected and entered.
  • The project is based on the “open” philosophy which also works in the area of open source software: the community is responsible for quality and completeness.
  • Do invite colleagues to join in and help - the more the better!
  • Keep informed and use the ’statistics’ option to follow the progress.

As stated before: it is an experiment. But we believe it will work - with your help!

What will this address information be used for?
Many applications and solutions in the area of geographic context need highly detailed data. Full address data including georeference is part of it.

[via: http://www.openaddresses.at]

Written and submitted from Home, using my 802.11g WiFi network.