-
- Lots of meeting planning!
- Posted reply to JFreeChart user forum to say thanks for advice and asked another question Point Symbols/Markers/Shapes
- Replied to an email from John Parsons who went to the Geospatial Mashup organised by Ordnance Survey to let CSAP members know that I am OGC Techincal Committee representative for the University of Leeds. As such I can gain them access to the materials on the OGC portal by setting them up with an account.
- Belinda has asked for an OGC portal account.
- MoSeS
- Discussions with Paul and Belinda about eResearch, Grid enabling and Grid HPC resources :)
- Portal Play Session and Design Discussion with Paul and Mark.
- This helped identify a few problems and improve our understanding of what the portal will be like for SC2006 and beyond.
-
-
- Joined JFreeChart user forum to try to get help... Answered a query that another user had which keeps a nice balance !)
- MoSeS
- Meeting with Paul
- There is a problem with how GeoTools and Tomcat are talking in the set up/query of some database.
- Paul is going to re-ask a question on the geotools user list that Pascal Neis was asking a few months back.
- Set some parallel jobs running...
-
- RBPAM seminar:
- Life on Earth: A Topographic Signiture of Life
- Mike Kirkby
- http://en.wikipedia.org/wiki/Daisyworld
- What are the suitable conditions for land based life?
- This may be very different for heat and pressure conditions such as those which have lead to ecosystems that are not based on solar radiation.
- Is plate techtonics being influenced by life?
- Much continental rock is composed of deadlife forms or their waste products that have built up in sediment.
- In addition to the density difference between mantle and crust material driving and currents in the mantle, plate techtonics is also perhaps driven by lubrication at subduction zones as a result of the more volatile nature of crustal material.
- Landscape at the local scale is massively effected by vegetation cover.
- Richard Dawkins once put forward a view that some of the first replicators were types of clay...
- Conway's game of life.
- Great! Geography branching out into the cosmos... :)
- Steven Pickering came and helped me test run some Parallel processing code (to test the PBS job scheduler on NGS).
- This confirmed there was still a problem for Steven.
- Hopefully this has got them to the next stage of fixing things.
- Going through this with Steven was useful for me as it has been a while since I did anything NGS wise :)
- The process was:
- Export my UK eScience certificate from my browser.
- Copy the exported certificate into a newly created .globus directory in my home directory on Maxima using WinSCP.
- On Maxima:
- cd .globus
- Convert the certificate as per user conversion on http://www.grid-support.ac.uk/content/view/63/42/
- openssl pkcs12 -in cert2007.pfx -clcerts -nokeys -out usercert.pem
- openssl pkcs12 -in cert2007.pfx -nocerts -out userkey.pem
- chmod 444 usercert.pem
- chmod 400 userkey.pem
- grid-proxy-init
- gsissh grid-compute
- Now on the NGS, run the following commands as per Steven's email
- module add clusteruser
- module unload pbs
- export PATH=/usr/local/Cluster-Apps/torque-2.1.6-snap.200610240247/bin:/usr/local/Cluster-Apps/torque-2.1.6-snap.200610240247/sbin:$PATH
- Copied and unpacked Stevens MPI programs from /tmp into my home directory /home/data01_b/ngs0381/
- qstat
- cd MPI/Examples
- qsub test.sub
- mpiexec was in the path but wasn't being picked up wierdly, so quickly fixed that.
- MoSeS
- Martin popped by for a check on progress :)
- Parallel processing
- Got furhter since upgrading to the latest Java5.0 (Build 9)...
- For some reason only first set of jobs processed...
- Bryan has some ideas about why it was crashing before... Adding some checking code...
-
- Helped Nanlin to run things on Unix :)
- Unfortunately the WUN GIS Seminar was postponed due to technical difficulties...
- MoSeS
- Parallel processing woes:
- Send... An unexpected error has been detected by HotSpot Virtual Machine:
- Suggestion is to upgrade JDK so have emailed Mike to sort this out...
- Aamir also suggested code checkpointing if I'm losing a lot of compute time.
- Checkpointing is a good idea... My first thoughts are that for results replicability I'll have to do some resets of the Random number generator, but that should be only a small addition and overhead...
- Paul has made progress with mapping using GeoTools :)
-
- Meeting with Martin
- Concerns over aggregated variables.
- Particularly Ethnicity and Economic Activity.
- General need to map and explore the census data.
- Good meeting although we spent the last part looking at the wrong table and I didn't realise until after.
- Realised that the denominators for Economic Activity should be age specific shortly after the meeting.
-
- http://portals.apache.org/
- The key to our future is collaboration!
- I woke up this morning thinking about a hill climbing dream I'd had.
- One thought I had was that I am trying to practice open research rather than e-Social Science.
- There is a distinction here with openness.
- The distinction also gave me more a focus on the Social Science aspect of e-Social Science.
- I thought about writing an essay on the topic, but decided instead to get on with some real work :-)
- MoSeS
- Meeting with Martin to discuss results. Not planned, but nevertheless welcome ;)
- There is a problem with the ethnicity graph... It was because I aggeregated the CASKS006 table wrongly.
- The plan is to produce graphs for Totals, as well as, reproduce the graphs for Proportions.
- I should get these to Martin and Mark in the next few days...
- Re-aggregated CASKS006 and produced a new Ethnicity Graph in Excel.
- Working out how to use JFreeChart...
- Phone call with Chris Higgins
- MoSeS/NCeSS eInfrastructure/OGC Grid Collison meeting in Edinburgh
- Alex Voss had to postpone the planned meeting this week for a month or so.
- In the mean time Chris will email to arrange a meeting with Paul Townend when he is up in Edinburgh.
- I will prepare with Paul for this meeting and attend virtually by telephone and email.
- OGC TC
- It would be good for OGC Grid Collision if I went to the next meeting as no one else can make it.
- I could take some instructions from Chris and be our ears and eyes.
- I'd like to do this, it is what JISC have given us some funding for.
- I should talk to Mark and if it is OK, then arrange flights and accomodation ASAP...
- Started planning to go to December 2006 OGC Technical Committee Meeting - San Diego, California
-
- MoSeS
- Paul asks what port is opened on the MoSeS portal server machine. I think it is 8080, but have said I'll check... It should be the case as long as Mike got round to it as he said he would... Not sure how we test this...
- Hmmm, some comm error problem means only have a few results from weekend... What went wrong this time?
- Andy Turner's Comparison_OAatMSOA_SWRHSARHPISARCEP Page
- Web Content updates
- Questions and queries
- Tim Motts is asking about GAM/K and the cluster software again. I explained that GAM/K is but one of the cluster hunting methods in the cluster software. I am sorry the development of this software has stalled, hopefully we can get things moving again one day... I asked Tim about his use of the software and whgat he is doing. Hopefully he will spur us on...
- Responded to Lynne Janbergs about University of Leeds OGC user list... I should ask and add Paul Townend to this...
- Responded to Swen Stoop's email with his DSOnline Gaming and Simulation Proposal.
-
-
-
- Web Content updates
- OGC Grid Collision / eInfrastructure / MoSeS
- Meeting with Mark to discuss this...
- MoSeS
- Emailed mpj-users list to try to get help with parallel programming
- Update Meetings Page
- Updated Software Page
- Downloaded JFreeChart.
- This is what Paul is using in the MoSeS portal.
- I plan to use it for the analysis of the population initialisation results.
- I was using Microsoft Excel, but should change before getting addicted...
-
- OGC Grid Collision / eInfrastructure
- MoSeS
- Trying to make parallel code more efficient
-
- Web Content updates
- OGC Grid Collision / eInfrastructure
- Need to get back to Chris about upcoming OGC TCs
- MoSeS
- Analyse results... Darn number of households is wrong... What is the bug... Or did I not run the right program?!... Haha, I'm losing every 28th result in the parallel version...
- MSOA optimisation too slow... Should the parameters be different?
- Mark needs me to document exactly what variables we are comparing...
-
- International Conference on Computational Science 2007
- Headsup from Mark about GeoComputation Workshop
- We should be involved...
- Start a Web Page to get things organised...
- Web Content updates
- MoSeS
- Sent a shapefile of Ward boundaries to Paul.
- There is no infomration about retirement in the Household SAR!
- Paul popped in for a chat :)
- SRDS was a success :)
- We talked about maps on the portal.
- Paul has a meeting with Mark at 11 am on Monday
- Paul reads my blog which makes me happy and justifies its existence. Long live e-Social-Science :)
- Oh Dear... Analysis reveals that I am not getting the number of households right in the population initialisation!
- Something was wrong in the optimisation.
- New results to be produced over the weekend... Fingers crossed!
- Mark needs me to document exactly what variables we are comparing...
- I really need to update Web Content especially with regard meetings and actions etc... I'm getting too far behind I might start losing important information...
-
- SoG Seminar by Mark. Nice one :)
- MoSeS
- Meeting with Mark and Martin.
- Andy to do a regression analysis of observed v expected for a number of variables:
- All_people
- All_households
- Each age category
- People_age_0-19 / All_people
- People_age_20-25 / All_people
- Etc...
- Unemployed_age_16-74 / All_people_age_16-74
- Retired_age_16-74 / All_people_age_16-74
- Male_economically_active_age_16-74 / All_people_age_16-74
- Female_economically_active_age_16-74 / All_people_age_16-74
- Ethnicity_white / All_people
- Households_with_No_Car_Ownership / All_households
-
- Sent Expenses claim for OGC meeting 2006-06-26 to 2006-06-30 to Chris at EDINA.
- Filed Expenses Claim for JISC meeting 2006-10-10.
- Linda emailed about Spatio-temporal Modelling in GIS Environments WUN Seminars. This looks great!) The web start seems like it will work for me...
- Web Content Updates
- MoSeS
- Aggregation of results for MSOA now sorted :)
- Trying to track down bug in MSOA runs...
-
- JISC First Joint Meeting of the GEMS/GEESE/GRID OGC Collision/NCeSS eInfrastructure Projects
- Participants as per agenda but with the following Subs:
- GEMS
- Keith Cole (presented) and Pascal Ekin
- Slides compiled from a number of projects:
- Data
- Neighbourhood Statistics and Census Data
- Need to talk to data providers to change licenses for data to be grid enabled?
- Workaround
- Shibboleth
- UK eScience certificate
- DN mapped to OGSA DAI interface
- Updated every 2 weeks
- Meta Data
- Usual OGSA DAI XML TableName etc...
- Undertaking work to parallelise some open source GIS software onto 128 cpu data node on the NGS
- GEESE/OGC Grid Collision
- Chris Higgins
- ISOTC211 and OGC are main Standards Defining Organisations for GIS
- OGC Grid Collision Programme
- Secure Access to Geospatial Services (SEE-GEO) Area A
- 3 Demonstrators
- NCeSS
- Use case and demonstrator MoSeS/GeoVue
- Using WCS for raster/gridded data not vector coverages?
- Semantically Aware Workflow Engines for Geospatial Web Service Orchestration (SAW-GEO) Area B
- Orchestration Area C
- Ordnance Survey Master Map (OSMM)
- JISC funding seperately
- Digimap style service
- SDSS team/NeSC/EDINA all closely located in Edinburgh
- OGF
- Enterprise Grid Alliance merged with the Global Grid Forum to form the Open Grid Forum.
- OGF20 meeting in manchester
- Opportunity to showcase OGC standards to OGF.
- SEEGRID
- GEOSS
- OGC is a member
- Attempt to get feedback from peers
- Detailed project plans for GEESE and OGC Grid Collision by mid November.
- Semantically Aware Workflow Engines for Geospatial Web Service Orchestration (SAW-GEO)
- Gobe Hobona working with Phil James, Paul Watson and David Fairbairn
- Now-2008
- Looking at WFS WMS WCS CSW
- Publish UDDI
- Standards
- Chaining services together
- Using Active BPEL
- Based on Eclipse plugin
- Current research phase to see what else is available
- Looking to collaborate with MoSeS - user testing and implementation
- NCeSS eInfrastructure
- Alex Voss
- Start Date 2006-12-01 (put back several months)
- GEMIDA
- Grid Enabled BHPS
- Metadata and markup-tagging of qualitative and quantitative data
- One of slides shows in italics the overlaps with other projects
- CAMS data
- Rob Proctor looking at how to virtualise this
- ISO19115
- 300 Metadata fields like dublin core?
- WFS
- GetCapabilities in OGC world is almost the same as WSDL bindings in Grid world.
- End of Talks
- David De Roure from OMII joined the meeting
- Taverna is going through OMII
- Workflow Languages
- SCUFL
- BPEL
- Scientific workflow and Business workflow are similar but communities are fairly isolated.
- Suitabilities and capabilitiesBPEL
- Discussion
- Chris needs to see an updated eInfrastructure proposal/workplan
- Alex has misread something so OGC-Grid collision Area C is not to do with training as he thought.
- Registries
- Catalog Services for the Web (CSW)
- INSPIRE
- Spatial Data Infrastructure (SDI)
- Richard is to be chair of OGSA DAI user group.
- In GEESE and OGC Grid Collision we want to use OGSA DAI to integrate spatial boundary data and census data.
- For MoSeS we also want to overlay contextual data using Google maps or OSMM.
- Web services Security v OGSA DAI security
- Review of security is in the first five months work of OGC Grid Collision
- Can OGSA DAI cope
- with large geographic data?
- with large numbers of users?
- Demonstrator deployment
- Need to keep an eye on GridShib and Shebangs
- JISC annual conference in March
- Meeting round up
- What priorities do we have?
- What do we need to keep momentum rolling? Another meeting - no, A wiki?
- References
-
- Officially this is holiday.
- Preperation for Brum tomorrow.
- Email Dan to suggest reading on Economic Geography etc...
- Email Peter Balmforth about EVIE, keep this in mind to suggest as a VRE for projects...
- MoSeS
- Network connection lost to BeoWulf... Processes halted and need restarting... done!
-
- Gillian is on to me to update the MoSeS NCeSS Web Pages...
- Provided expertise to Stephanie Garside and Martin Clarke about spatial look up tables from census to postcode areas and vicea versa.
- PhD
- Oli emailed a recap of the Research Suggestion from our meeting yesterday:
- Using CAS, you can determine the numbers of people in various occuptations and industries etc at a fine scale (e.g. OAs)
- Getting the target 'daytime locations' is harder. Look at ST theme table TT010, which has daytme and workplace populations, although it's only at ward scale. I'd don't think that there is any OA level occupational data; this would almost certainly be disclosive though, so I'm not surprised.
- Create a model with known transport infrastructure.
- Stage 1: agent model the people in diff occupations at the residences, and get them to go to work, such that you end up with the right number of people in different occupations at the workplace end. What patterns do you get?
- Stage 2: You can improve this by using the SWS, which tells you where (in aggregate) the people living in a given location go to work. Are your agents in stage 1 going to the 'right' locations? If not: are they going to an economically/environmentally/rational 'better' location? Or is your model simply wrong! :-)
- Stage 3: You can improve it further on the basis that SWS3 gives you (SCAMmed) OA to OA flows by transport mode. Are your agents using the 'right' transport? etc...
- If you succeed in getting agents to go to the right locations using the right transport: a) well done! b) try changing the transport infrastructure.
- Somewhere in all of this, compare results to a standard stocks based spatial interaction model.
-
- Talk with Lindsay Banin about visualisation of point data.
- Neighbourhood Knowledge Management Workshop
- EGC meeting :)
- MASS meeting
- Meeting with Oli
- Spatial interaction data
- MoSeS
- PhD
- A model of daily activity:
- Start off with everybody at a home or residence by default
- Each person is an agent and they choose by random an available transport mode to work
- Incorporate journey to work
- Move people from home to work and back again on a 15 minute time step
- Each day a new transport mode is chosen based on the weather and experience of previous journeys
- Try to settle things down to get an average journey mode profile for an area
- Compare this with 1991 SWS mode of transport data
- Incorporate a new journey, e.g. journeys to school and re-examine
- Oli kindly agreed to help support me in this research so hopefully he will join my RSG. I also hope to have Haibo and Gordon on this :)
-
- MoSeS
- Meeting with MB, MC, HC, JK, BW
- CSAP launch
- DSOnline Gaming and Simulation telecon
-
- Tim Mots user query about Cluster Hunter.
- MoSeS
- I think I've worked out why I was struggling with the parallel processing:
- Using MPI.ANY_SOURCE instead of a specific source in recv method...
- Read IMA Newsletter of October 2006
- Passed on to CSAP
- Should check out http://www.ssb.no/misi/ especially Simon Gault, Department for Work and Pensions, United Kingdom: A Generic Approach to Dynamic Microsimulation Modelling (GENESIS) work.
- Following discussion on SDIs on the OpenSDI email list...
- Links
-
- MoSeS
- I think I've worked out why I was struggling with the parallel processing:
- Using MPI.ANY_SOURCE instead of a specific source in recv method...
- Read IMA Newsletter of October 2006
- Passed on to CSAP
- Should check out http://www.ssb.no/misi/ especially Simon Gault, Department for Work and Pensions, United Kingdom: A Generic Approach to Dynamic Microsimulation Modelling (GENESIS) work.
- Following discussion on SDIs on the OpenSDI email list...
- Links
-
- Manuel emailed a first draft of the fire paper that I helped him format some data for... Had a quick flick through and it looks worth having a closer read. Manuel wants some feedback...
- Read
- DSOnline Gaming and Simulation
- MoSeS
- Need to do a progress report for the last month by the end of tomorrow
- Programming
- Get going with Sampling with replacement...
![[Validate RDF]](http://www.geog.leeds.ac.uk/people/a.turner/images/rdf_w3c_button.gif "W3C RDF Validation")
![[Validate CSS]](http://www.geog.leeds.ac.uk/people/a.turner/images/vcss.gif "W3C CSS Validation")
![[An image of Andy Turner]](http://www.geog.leeds.ac.uk/people/a.turner/a.turner.png)