We're sorry but this page doesn't work properly without JavaScript enabled. Please enable it to continue.
Feedback

Gflowiz: Flow maps in the geoweb

00:00

Formal Metadata

Title
Gflowiz: Flow maps in the geoweb
Alternative Title
gFlowiz, an open science framework to analyze and geovisualize networks and flow datasets
Title of Series
Number of Parts
295
Author
Contributors
License
CC Attribution 3.0 Germany:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
Language

Content Metadata

Subject Area
Genre
Abstract
Geographic flow visualization (gFlowiz) is an interdisciplinary project dedicated to flows and networks in the geoweb. It is led by a team of French researchers and engineers in Geography, Cartography and Computer Science from both IFSTTAR and CNRS. gFlowiz follows the new paradigm of « visualization mapping » (MacEachren, 2004) not present in current geoweb applications. All types of flows, movements and traffic in a geographical space, at several scales, are involved (migrations, freight, transportation, etc). Required semiological, aesthetical and design elements offered by currently web-based graphic visualization libraries are also considered. Its main goal is to build a simple application combining two pillars of scientific graphical representation: flow data processing and (geo)graphical modalities visualization. A state of the art on current issues of flows and movement analysis on the geoweb has been produced through the compilation of around 70 applications in a thematic dashboard, and a 200 respondants survey on flow map usages and needs has been realized. This presentation will follow 3 axes: open source applications in the curated corpus, an analysis of the survey and a current development status of the application
Keywords
Computer clusterComputer networkLaurent seriesPresentation of a groupComputer animation
RobotProjective planeDifferent (Kate Ryan album)Projektive GeometrieMappingMereologyLecture/Conference
Archaeological field surveyComponent-based software engineeringVisualization (computer graphics)Web applicationVariable (mathematics)Addressing modeTexture mappingInformationNeuroinformatikOpen sourceVisualization (computer graphics)Web applicationData managementArchaeological field surveyOnline helpState of matterProjective planeMachine visionSoftwareUniform resource locatorMappingComputer scienceQuicksortGoodness of fitLibrary (computing)Open setCartesian coordinate systemVariable (mathematics)Point (geometry)Component-based software engineeringOperator (mathematics)Position operatorBitAdditionWebsiteWeb 2.0RoboticsTransportation theory (mathematics)Traffic reportingWhiteboardUltraviolet photoelectron spectroscopyLecture/Conference
Frame problemFile formatArchaeological field surveyPrice indexCartesian coordinate systemCommutatorDifferent (Kate Ryan album)Group actionTexture mappingMultiplication signTransportation theory (mathematics)Goodness of fitZirkulation <Strömungsmechanik>Mobile WebSimulationTerm (mathematics)SoftwareWeb applicationFreewarePoint (geometry)Category of beingContext awarenessSemantics (computer science)GeomaticsWeb 2.0System administratorComputer fileGeodesicFamilyOpen sourceDependent and independent variablesScaling (geometry)State of matterService (economics)Open setCASE <Informatik>Archaeological field surveyMathematicsInformationMappingVector spaceValidity (statistics)Position operatorProjektive Geometrie40 (number)WebsiteRegulator geneWordDiscrete element methodInterpreter (computing)Local ringParameter (computer programming)
Archaeological field surveyEndliche ModelltheorieData modelSoftware testingBeta functionLevel (video gaming)Open setUniqueness quantificationMappingCombinational logicPoint (geometry)InformationOpen setOpen sourceSoftwareWeb applicationCartesian coordinate systemDistanceMathematicsForm (programming)Texture mappingComputer scienceLevel (video gaming)Category of beingGoodness of fitTime zoneProjektive GeometriePosition operatorMereology1 (number)Student's t-testWebsiteTunisBitUniqueness quantificationLink (knot theory)Arrow of timeFreewareFilter <Stochastik>Web 2.0StatisticsSoftware testingDefault (computer science)CountingRepresentation (politics)TriangleGraph coloringServer (computing)CodePlotterVector spaceGraph (mathematics)Closed setEndliche ModelltheorieComputer animation
Transcript: English(auto-generated)
Okay, thank you. Welcome to the newcomers. I will start the next presentation. So, I will speak about the JFrogovice project, which is a project on flow maps in the JWO web.
I will speak about the project globally and I will speak about the three parts that composed that project. The corpus we gathered, a survey we made and the arabesque web application we are currently developing.
So there is three components in that project. We gathered information on current flow visualization web apps and we gathered it in the corpus. And we studied it a little bit. So I will show you what data we extracted from the corpus.
I will speak about the survey we made to get insight about what are people expecting on flow maps application.
And I will speak about the arabesque web app. So this project is developed by a team of researchers and engineers in geography, computer science and computer vision. And from the IFSTA French Institute of Research on transportation, planning, management and
networks and the CNRS, which is a multi-purpose national research center in France. And we had the help of two interns during the summer last year and this summer.
So first of all I will speak about the corpus. So we gathered information about 79 web applications available on flow maps and this kind of sort of application. It's about 54 variables so it's a pretty exhaustive review. The purpose of that was to get
a kind of state of the art of what is available on web application for doing flow maps. And we gathered it in a dashboard. We developed it with Rshiny which is already available at this URL.
So it looks like that. You can see the raw data. You can extract it with a CSV or xFi and you can also access to a dashboard to see some graphics about the data.
You can also filter the data if you want to figure out about it. So it's already available. In that corpus I extracted maybe seven variables and then we'll speak about it. First one I want to show you is that about the licenses.
Most of the application we don't have information about the license but for the one we get the information we can see that there is a lot of them that are open source or lib. And a bunch of them are proprietary. So it's a good point for open source software.
An example I can show you maybe it's an open highway map which is a map from open street map data but on the highway data which is collaborative and open source. And the other side we have Strava which is a property application to see applet parkours, runners and such.
So we have this kind of application. We get data. We take snapshots of it. The context where these applications are developed are often the private sector by private companies and such.
There is a bunch of work from the public sector and the research. So if we gather them it's a big group. Some of them are collaborative like the one I showed you before. And some of them are from personal professionals.
So it's professionals that are doing that on our free time. One example I can show you is the name NIM and ALES Urban Agency which are cities in the south of France who published web application to show the commutes between the cities at different times of the day.
And we have the GoBike Trade Africa which is a web application to visualize goods flows between Africa and mostly you can see it but it's Africa and China.
The thematics used in that application are mostly about transport, migration, networks and mobility and others. This is the most terms that are used in the application. For example this is transportation data.
It's a simulation of the transport services in Switzerland from GTFS data set.
So it's both services and the quantity of them in Switzerland. The different scales. Most applications use global scale. Some are national and regional or local scale. Very few of them are multiscale.
We are developing our best web application which is multiscale in this case. We give information on features provided by this application. Some of them are very common like zooming, panning, adding vector data.
You can filter data often. And some of them are less common like projection, transparency, handling, change administrative layout. And as you can see we implemented common features but also not so common features in our basic application I will show you after.
You can see that it's web apps so most of them use JSON and the JSON family of data formats.
A few of them use CSV. We choose to use CSV files for the flow maps and some of geodesism for the nodes. You can add geodesism for the nodes in the application we developed. We also made a survey to see if we have a state of the art about what people are expecting from a new application.
What would be interesting to have in the application. So we got 200 respondents to this survey. It was made in France. So it was in French.
Most of them are from the public sector. 80% are from the public sector. 30% were from the research sector. So maybe it's a bias from the people we were able to touch but maybe it's our community as we are from the research side.
And things that are interesting is that the three important tools we are using, two of them are free open source software. So mostly we use QGIS and we also use R to visualize geospatial data and we also use ArcGIS.
So it's a good point for free open source software for geomatics. The semantics are similar to what we saw before. It was about people transportation and goods transportation mostly.
Some of them are interesting like ideas circulation and it's mostly how ideas circulate between researchers. And you can see some people working on mobile phones or ticketing technology and such.
Maybe it's still a bias from the community we reached but people wanted a free application and open source application if possible. It was the most asked features even before an aesthetically pleasing representation or an application easy to use or free.
So we asked them what do you want as a graphic model. Mostly they asked floor maps.
Some of them were asked for corporate maps. We can do floor maps in the web apps we developed, not corporate maps. It's a feature we will add. We can do a statistical chart, simple ones like bar plots and such.
But we won't do network maps, graph or vector field. This is not the goal of our application. So we developed an application we called Arabesque which is a dance move.
It's a web application. It's developed with robust technologies like OpenLayers, D3.js and WebGL, Porch4.js to handle the projection part. It's still in beta testing stage. The beta testing stage is not open yet.
And it's developed by one of your interns which is an engineer and computer science student. It is in business to close license so it's open source even if the code source is not on the depot which is public.
So I can't show you a demo yet. It's not on a public server yet too. But I can show you a snapshot we did. The application is in three parts, two panels and a map. On the left
panel you can play with your layer, add new layer, transform them to another projection. There is currently around 15 projection available and we plan to add more. You can add layers from GeoGzone or from OpenStreetMap.
And you can order and change the color scale, change the form of the arrows and such, or the link. On the right panel you can add filters. You can filter your data from information provided by your data.
So maybe from the distance between each link which is calculated automatically or the count of your nodes. You can add your filter as well. This is a filter provided by default but you can add your own filter.
Maybe you can add categorical information you want to filter. And you can add one or two of the side panels to show the map. So this is one of the kind of representation you can do like it's a curved triangle.
And you can add also when you go over a feature you can get information about the feature. The final release will take part in the Sergio conference in France in November this year.
I hope we saw you there to get this information. To conclude, we use robust and well-known technologies to do that.
Because we knew it would be easier and we don't get too much problem with it. It's focused on flow maps. For now it's just flow maps but it does as well.
It's a unique combination of features as we showed before. And we pay a great attention on graphics. So the graphics are pretty and you can have pretty and nice maps with it. And last point, I like it freer and faster software.
Thank you for your attention.