Next Generation Web Mapping Webbased Mapping for the
Next Generation Web Mapping Web-based Mapping for the 21 st Century Archaeologist Bill Dickinson Jr. GEOG 596 A, Spring 2007 Penn State World Campus MGIS Program
Overview • • • Evolution of Web-based Mapping Next Generation Mapping An Archaeological Application 2 D Mapping Client 3 D Mapping Client Conclusion
The Evolution of Web-based Mapping
Previous Generation • Client technology Client Server “Dumb” map Limited spatial capabilities. – Static map image with single level of detail exposed as flat image file on web page. – User interface features limited to form-based HTML information requests. – Geospatial toolset limited to moving around flat image within single level of detail (pan). • “Smart” mapping application Broad spatial capabilities. Server technology – Map images generated on-the-fly and temporarily stored as a single image file. Changes in level of detail required generating a new map image. – Client-server communication through complex HTTPRequest strings passed to map server through web forms. – Limited spatial capabilities accessed through restrictive and proprietary XML variants (Arc. XML and GML). – Arc. IMS 4. 0, Web Mapping Server (WMS).
Current Generation • Client technology Client Server “Dumb” map Limited spatial capabilities. – Dynamically generated map image with multiple levels of detail exposed as flat image file on web page. – User interface features primarily graphical in nature (I. e. , buttons) that activate associated application actions. – Geospatial toolset expanded to navigating through the map (zoom in, zoom out, pan) as well as a small set of geoprocesses available to the user (query, geocode, buffer). • Server technology “Smart” mapping application Broad spatial capabilities. – Map images still generated on-the-fly and temporarily stored as a single image file. – Client-server communication expanded to include ASP -postback in. NET web applications. – Expanded spatial capabilities accessed through restrictive and proprietary XML variants (Arc. XML and GML). – Arc. IMS 8. x/9. x, Arc. GIS Server 9. 0 -9. 1, Web Feature Server (WFS).
Next Generation • Client Server “Dumb” map Limited spatial capabilities. • “Smart” mapping application Broad spatial capabilities. Client technology – Highly interactive, dynamically generated map views with unlimited levels of detail exposed as tiled image caches. – User interface features are fluid and generated on an as-needed basis depending on user actions, closely mimicking traditional desktop mapping clients in responsiveness and scope of features. – Geospatial toolset expanded to full suite of traditional desktop mapping applications (layer manipulation, attribute manipulation, surface analysis, 3 D analysis, etc. ) as well as unique, highlycustomized geospatial tools for specific user groups. Server technology – Map images stored as pre-generated, tiled map caches at multiple levels of detail for fast retrieval and transfer to web client. – Client-server communication through SOAP, REST, or RPC calls. – Broad spatial capabilities (including 3 D visualization) exposed as web service end-points using a flexible geospatial XML variant (KML). – Arc. GIS Server 9. 2, Arc. Web Services, Google. Earth Server, Space. Time Mapping Server.
Next Generation Mapping
Next. Gen Map Applications • New client technologies will provide multiple users with an experience very similar to a traditional desktop client. • New server-side approaches will provide for extremely fast data transfers of map imagery for both 2 D and 3 D client interfaces. • New geospatial web services will provide for a very large geospatial toolset that could be consumed by many different client technologies in whatever combination desired by the individual user.
Service Oriented Architecture There is no widely-agreed upon definition of serviceoriented architecture (SOA) other than its literal translation that it is an architecture that relies on service-orientation as its fundamental design principle. Service-orientation describes an architecture that uses loosely coupled services to support the requirements of business processes and users. Resources on a network in an SOA environment are made available as independent services that can be accessed without knowledge of their underlying platform implementation. -- Wikipedia, 17 May 2007
Web Mapping Architecture • Geospatial tools are developed as loosely coupled, independent web service packages. • Mapping applications are accessed through standard web technologies to meet the needs of a specific business process or user. Clients KML/ SOAP Web Services KML/ SOAP Data Stores
Benefits of SOA • Allows for distributed, web-based mapping applications that can be rapidly engineered for each unique need. • Accessible to the widest possible user-base with the smallest application footprint possible on the local computer. • Connects light-weight mapping client applications to relevant geospatial and legacy data stores. • Leverage server-side geoprocessing tools for distributed analysis.
An Archaeological Application
The North Orkney Historic Population Project • Examine changes in demography, economy, settlement, and land-use in the north Orkney Islands (Westray, Papa Westray, Faray, Eday, Sanday, and North Ronaldsay). • Period of interest is c. AD 1735 -2000, which witnessed two transition trends: – From traditional subsistence production based on small-scale farming and fishing to modern system of beef production for external markets. – From sustainable demographic regime with limited migration to system with substantial out-migration, depopulation, and population aging. • Project intended to study the spatial relationship between the factors in an attempt to better understand the forces of demographic change in a relatively contained population.
Project Assumptions • New models of demographic change should be explicitly spatial in nature. – “Homeostatic demographic regimes” (a population in equilibrium sorted by clusters of habitats from which necessary resources are drawn) have both temporal and spatial dimensions. – Modern demographic change always involves a significant reorganization of population across space. • Remotely sensed imagery should provide new perspectives on the environment of the study population. – The northern islands of Orkney are virtually treeless, providing near-optimal conditions for the use of remote sensing as an environmental and archaeological survey method.
Identified Mapping Tasks • Geo-reference all house sites(buildings) and link them to households (people). • Link existing information on topography and soils to new data taken from aerial photography and satellite imagery. • Digitize and geo-reference cadastral maps, rentals, tax valuations, and other old documents kept onisland or in the Orkney Archives (Kirkwall, Orkney). • Test new demographic theories in a spatially heterogeneous environment. • Link all data in a single GIS system.
Map Application Requirements • Use Arc. GIS Server as data store and geoprocessing backend. • Combine project raster (satellite imagery and scanned paper maps) and vector (points, lines, polygons) data into single interface. • Provide access to specialized project “business logic” (i. e. , demographic change analysis tool).
Applying Next. Gen Mapping • Next. Gen web-based mapping applications are built on a “loosely coupled” architecture, which makes it ideal for combining disparate functions into a single user interface. – Arc. GIS Server can expose map services using various “open” methods used in Next. Gen mapping applications. – Next. Gen applications use standard internet graphics files to display both raster and vector data in the user’s web browser of choice. – Specialized business logic is accessed through custom web services hosted on the server. • Additional benefits: – Allow end-user’s to access project data independent of computer platform or location. – Provide a means for users to collaborate even if separated by large geographical distances. – Use a client interface (a web browser) that is nearly ubiquitous in the world today.
2 D Mapping Client
Adobe Flex • 2 D Streaming Map Application • Flash-based technology – Action. Script 3. 0. – Vector graphics engine. – Embedded in modern browsers with plug-in. • Enhanced with ESRI’s Flex SDK for rapid application development. • Uses Arc. GIS Server 9. 2 for backend geoprocessing tasks and data storage. • Leverage ESRI’s new Arc. Web Services (AWS) tools. – http: //www 2. arcwebservices. com/v 2006/index. jsp – Access data from more than 20 leading commercial mapping data and content providers. – Access standard geoprocessing web services (geocoding, queries, find-a -place, and more).
Basic Tools • Basic map application tools: – Spatial navigation (zoom in, zoom out, pan, full extent, zoom to layer, zoom to feature, spatial bookmarks, etc. ). – Spatial analysis (buffer, query, identify, geocode, etc. ). – Spatial editing (create/delete/move/modify feature or records). • ESRI’s generic Flex interface can be found at http: //www. arcwebservices. com/awx/index. jsp. – Shows basic toolset as well as gives feel for how Flex mapping client will perform.
Advanced Tools • Embed streaming media. – Streaming audio and video can be easily embedded in a Flex application. – Audio interviews can be accessed geospatially through map interface, given analysts ability to review audio records within the application. – http: //www. amgentourofcalifornia. com/docroot/tourtracker 2/index. html • User collaboration. – By adding Adobe’s Flex Data Services (FDS) on the server, dynamic user collaboration can be achieved through the mapping application. – Collaborative features can include: • Single web user controlling the application experience for multiple users. • Editing of data by one user can automatically be pushed through to all other users.
Project Tools • Specialized project tools: – Population well-being algorithm exposed as web service. • Will allow for dynamic input of algorithm parameters so that analyst can “tweak” the output depending on various factors. – Timeline viewer. • Will allow analyst to dynamically apply a time filter to the spatial data to look for spatiotemporal patterns in the underlying data. • Data can be “played” through time (backward and forward) for further analysis. Simile Timeline, a DHTML-based AJAX widget for visualizing time-based events. (http: //simile. mit. edu/timeline).
Flex Pros and Cons • Benefits – Streaming application means minimum features are downloaded immediately with the bulk of the application loading in the background. – Vector engine will allow for smoother vector data layers to be drawn internal to the client. – Vector data can be retrieved from server as simple XML feed and then converted to 2 D features locally, reducing the data transfer load and increasing the responsiveness of the map application. – Can take advantage of local computers graphics card for data visualization. – Action. Script 3. 0 is a full-featured Object-Oriented (OO) programming language and will provide for more intensive geoprocesses to be built directly into the client. • Drawbacks – Requires browser plug-in, which can be a deterrence to some users. This is offset by the fact that all modern web browsers come with the Flash player pre-installed.
3 D Mapping Client
3 D Visualization • 3 D is truly the next generation of mapping applications. • Will provide new ways to analyze data not available with 2 D “flat” maps. • Potential 3 D client options: – Flex 3 D – Google. Earth – Space. Time Java 3 D Client – Arc. Explorer (AGX)
Flex 3 D • Advantages – Would be internal to web-based client. – Best performance for vector-based 3 D due to Flash graphics engine. • Disadvantages – Unproven technology that would potentially require significant development time to realize. • Would have to convert a Flash 3 D library into something that could be used in a Flex mapping application. – Sandy 3 D Flash Library (http: //www. flashsandy. org/). – Away 3 D Engine (http: //away. kiev. ua/away 3 d/). – Would have to limit 3 D experience to vector representations as the use of raster data is not viable yet.
Google. Earth • Advantages – Basic viewer is free. – Best mapping 3 D data visualization tool currently on the market. – Arc. GIS Server can publish data as KML services, which are consumed natively by Google. Earth client. – Free access to current high resolution satellite imagery for nearly all parts of the world. – Development path for creating 3 D models of real world objects. • Disadvantages – Not in web browser - need to download stand-along client. – Useful tools are only available with purchase of most expensive client license ($400). – Closed API - developing customized extensions is a very expensive route.
Space. Time Java 3 D Client • Advantages – Agile 3 D analysis and display environment. – Deployed in modern web browsers. – Will run on any platform with Java 3 D libraries installed on system. – Built-in support for spatiotemporal data visualization through Geo. RSS feeds. – Easily extensible for customized geoprocessing needs through exposed web services. • Disadvantages – Requires Java 3 D libraries to be installed and configured on local computer. – Potentially longer development time to embed desired tools within client.
ESRI Arc. GIS Explorer • Advantages – Best full-featured GIS 3 D mapping client currently on market. – Geoprocessing capabilities easily extended with complete suite of ESRI geoprocessing tools as seen in Arc. Map. – Can connect directly to Arc. Catalog data source. • Disadvantages – “Thick” client - completely separate, stand-alone mapping application. – Requires expensive user licenses. – Not web-based, though it can connect to remote data sources through the HTTP protocol. – Performance is not as good as other options. – Requires a lot of local computer resources (CPU, RAM) for best performance.
Conclusion • Outcome of proposed project: – A Next Generation web mapping application using a Service Oriented Architecture. – A 2 D web client built on Adobe Flex 2. 0 and ESRI’s Arc. Web Services SDK. – A 3 D web client using the Space. Time Java 3 D web client. – Specialized business logic exposed as independent web services using the SOAP communication protocol. • Questions?
- Slides: 30