Andy Turner's Blog 2009-08 Web Page @ School of Geography, University of Leeds

Blog
Events:
- Geomorphometry 2009
  - http://geomorphometry.org/content/geomorphometry-2009-programme
  - 2009-08-31 to 2009-09-02, Zurich, Switzerland
- Workshop on The impact and influence of Web 2.0 on e-Research Infrastructure, Services and Applications
  - Held in conjunction with the IEEE International Conference on Cluster Computing (Cluster 2009)
  - 2009-08-29 to 2009-09-04, New Orleans, Louisiana, USA
- Events on the horizon:
  - GIScience 2010
    - http://www.giscience2010.org/
    - 2010-09-14 to 2010-09-17, Zurich, Switzerland.
  - ComputationWorld 2009
    - http://www.iaria.org/conferences2009/ComputationWorld09.html
    - 2009-11-15 to 2009-11-20, Athens, Greece.
  - NGS Innovation Forum '09
    - 2009-10-21 to 2009-10-22, London, UK
    - http://www.ngs.ac.uk/innovationforum09
  - Workshop on The impact and influence of Web 2.0 on e-Research
    - 2009-10-13, Banff, Canada.
    - http://research3.org/events/workshop_two.php
    - Held in conjunction with OGF27 2009-10-12 to 2009-10-16
  - European Conference on Complex Systems 2009 (ECCS'09)
    - http://eccs09.info/
    - http://www2.warwick.ac.uk/fac/cross_fac/comcom/events/eccs09/
    - 2009-09-21 to 2009-09-25, Warwick, UK
  - The 6th European Social Simulation Association Conference
    - http://cress.soc.surrey.ac.uk/essa2009/
    - 2009-09-14 to 2009-09-18, Guilford, UK
  - Geomorphometry 2009
    - http://geomorphometry.org/content/geomorphometry-2009-programme
    - 2009-08-31 to 2009-09-02, Zurich, Switzerland
2009-09-02
2009-08-28
- Browsing
- Miscellanea
  - PC crash
    - Nightmare :-(
- e-Science
  - GENESIS
    - Social Simulation
      - Leeds Model Development
        
        Obtained and wrote loaders for UK 2001 Human Population Census Special Workspace Statistics (SWS) data at Output Area (OA) level
        
        Obtained and wrote loaders for OA Easting and Northing data
        
        ...
      - Attempting some small city scale simulations:
        
        10000 Households, 2 people in each, region 4096 rows 4096 columns
        
        Incrementing every minute for 1 week
        
        After 13 ticks this slows down dramatically probably due to memory reasons...
        
        Trying 1000 Households...
2009-08-27
- Browsing
- e-Science
  - GENESIS
    - Social Simulation
      - Attempting some small city scale simulations:
        
        10000 Households, 2 people in each, region 4096 rows 4096 columns
        
        Incrementing every minute for 1 week
        
        Purpose of these is to test the code further, test for scaling issues, and see what output is like (visualation issues abound...)
        
        ...
      - Developing network and movement code...
        
        Second version is now implemented and available to the world
        
        Default in the first version movement was either the diagonal part first, then the vertical or horizontal part next, or vice versa.
        
        Default in this version is the shortest straightest path.
        
        People now develop a history or record of the movements they have done as a network.
        
        People are able to travel at different speeds.
        
        Visualisation has been improved by displaying networks over rasters.
        
        Example outputs
2009-08-26
- Browsing
  - http://www.earthsystemgrid.org/
  - http://www.giscience2010.org/
- e-Science
  - e-ISS
    - Had a look at management meeting notes from today:
      - http://portal.ncess.ac.uk/access/wiki/site/e-iss/2009-08-26.html
  - NeSS
    - Meeting with Junaid
      - GLS
        
        Working with Michael Koutroumpas
        
        We found a Sakai Tool called "Portlet Edina" in the list of available tools on the NCeSS Sakai Portal and added it to various sites.
        
        The GDAS URL for census data that we are to use is:
        
        http://census.mimas.ac.uk:8080/gdas/gdas
        
        The NCeSS VM Server URL is:
        
        http://husserl.ncess.ac.uk:8080/proxy/Proxy
        
        The NGS VM Server URL is:
        
        http://gls.vidar.ngs.manchester.ac.uk:8080/proxy/Proxy
        
        We did not get these to work although Junaid had had more success yesterday...
        
        We tried a wget of the request from gls.vidar.ngs.manchester.ac.uk:
        
        wget "http://gls.vidar.ngs.manchester.ac.uk:8080/proxy/Proxy?Service=WPS&version=0.4.0&Request=Execute&Identifier=Geolink&store=false&DataInputs=AccessMechanism=OGR:SHAPE;GeolinkedDataHost=http://census.mimas.ac.uk:8080/gdas/gdas;GetDataVersion=0.9.1;FrameworkDomain=edina.ac.uk;FrameworkName=english_oa_2001;FrameworkVersion=2001;DatasetDomain=census.ac.uk;DatasetName=2001CAS;DatasetVersion=2001;DatasetGeolinkageField=ons_label;Attributes=ks0080001;GeolinkIds=00DAGH0065"
        
        Result was connection refused
      - Junaid successfully signed off from Security Work Package
      - Working with Paul to get Sakai test install for migrating MoSeS portlets
      - Paul hopefully still working on a draft for demonstrator funding or has liaised with Mark
        
        Andy to ask Mark about this later today...
  - GENESIS
    - Meeting with Mark
      - Gave a demo of the GenerateSociety_5
        
        Still some bugs to sort out...
      - Mark pointed out that there is work to do comparing how I have done networking/routing/agent movement and how others have...
    - Social Simulation
      - Developing network and movement code...
        
        ...
2009-08-25
- Browsing
- Miscellanea
  - Geomorphometry
    - Completed survey:
      - http://geomorphometry.org/content/geomorphometry-society-meeting-e-democracy
- e-Science
  - GENESIS
    - Social Simulation
      - Developing network and movement code...
        
        ...
2009-08-24
- Browsing
- Miscellanea
  - Road Accident Data Usage
    - Helped Zhuyu Zhang, an MSc student at Imperial Collage London by registering study number 40885 with ESDS for accessing Stats 19 data.
    - http://www.geog.leeds.ac.uk/people/a.turner/data/Stats19/
- e-Science
  - GENESIS
    - Social Simulation
      - Developing network and movement code...
        
        Fixed logic error in intersection code...
2009-08-07
- Browsing
- e-Science
  - GENESIS
    - Social Simulation
      - Developing network and movement code...
2009-08-06
- Browsing
- Miscellanea
  - Completed questionnaire as part of a Research Information Network (http://www.rin.ac.uk/) survey to study changes in communication and publication practices by researchers.
    - http://www.ncess.ac.uk/research/hub_research/useandresearchofweb2/
    - Survey Link
- e-Science
  - GENESIS
    - Social Simulation
      - Developing network and movement code...
2009-08-05
- Browsing
- e-Science
  - GENESIS
    - Social Simulation
      - Developing network and movement code...
      - In any raster network, no matter what the detail in the network connections, if these also follow a regulr pattern there are at least two equal distant ways to get from one point to another
        
        This means at some stage a decision must be made about what path to traverse...
        
        It also means that there is little benefit in adding much detail beyond neighbouring points in the raster...
      - Movements
        
        On day one each person sets off on a shortest and straightest first path to work.
        
        Their network evolves and it is easier for them to go back a route they have learned than it is to go back another way.
        
        A network route that is learned is not as easy to travel as one that they have travelled, but it is easier to travel than one they know nothing about.
        
        Generally the network evolves so that the more people use it, the more people it can carry (and faster) upto limits which when exceeded, its flow slows.
        
        Exactly how this feedback works I'm not sure...
        
        It seems sensible as a start to constrain networks...
        
        There could be a big difference between the evolution of networks that start with nothing and involve no additional journeys than those that already start with some detail.
        
        It will take time even for an unconstrained network to settle down for a regular set of journeys made to and from work in a synthetic society as there is feedback.
        
        To start with the only transport mode is walking, but we want to have cars and buses...
        
        People are to learn about journeys of others and with others they meet at work they may plan to share parts of their journey.
2009-08-04
- Browsing
- e-Science
  - NeSS
    - Meeting with Paul and Junaid about demonstrator funding
      - We agreed to propose something along the lines of:
        
        Make the MoSeS system available in the NCeSS Sakai Portal
        
        Integrate ConvertGrid so that output can be aggregated to different geographical areas
      - At some stage soon we should make Pascal Ekin aware of our plans so that he can give us some feedback on our use of ConvertGrid.
      - This will fund Junaid until the end of the project and Paul willing to divert GENESIS effort into this for a month.
      - Paul to make a first draft of the proposal..
      - Sakai
        
        Paul keeps having trouble installing Sakai which we want in order to test moving the MoSeS system across.
        
        Junaid is going to help with his experience of installing Sakai...
        
        Plan b is that I contact Rob Allan to get some testbed set up at Daresbury.
        
        We are unsure, but suspect that Steve Swinsburg is no longer working on the project...
        
        Andy will check with June Finch
  - GENESIS
    - Social Simulation
      - Developing network and movement code...
2009-08-03
- Browsing
- e-Science
  - GENESIS
    - Social Simulation
      - Developing network and movement code...
2009-07-31