-
- Browsing
- e-Science
- White Rose Grid e-Science Centre Workshop: e-Science Challenges and Perspectives
- MoSeS
- References
-
- Browsing
- RLF Meeting with Linda See
- This meeting came about because of an enquiry by a graduate Guo Xiaofu who was reviewing RLF/1 and found that the online information and system were not there (See 1).
- This is because the CCG Web Server machine died and it was not a simple task to replace it.
- River Level Forecaster (RLF/1) was developed by Simon Corne in the late 1990s.
- Linda says that there was a lady based in Engineering that used RLF/1 in teaching and we wonder if Guo was a student of hers?
- Anyway, the idea of this meeting was to try to ascertain where all the files were and what they are with a view to outlining a Masters Dissertation project to get RLF/1 back up and running again.
- RLF/1 was primarily written in C++. There was both a web interface and a command line interface.
- The main run file for RLF1 is on a drive called the "N" drive at the following location:
- In some documentation it is clear that RLF/1 is built on the Solaris 2.5 Operating System.
- We have not investigated what of the operating system is used in the compilation and running of RLF/1.
- The old URL: http://www.ccg.leeds.ac.uk/simon/intro.html does not work, but thanks to The Wayback Machine there is a copy:
- A complete archive of web pages and is on the "N" drive at the following location:
- References that might be useful:
- CCG item: http://www.geog.leeds.ac.uk/people/a.turner/personal/blog/archive/2008/03/#2008-03-10
- Jennings, E., 'An investigation into the application of artificial neural networks for flood forecasting in the Ouse catchment, Yorkshire'. MSc thesis. University of Leeds. 1998.
- Corne, S.A. (2002) RLF/1: a World Wide Web Flood Forecaster. In R. J. Abrahart, L. See & P. E. Kneale, eds., chapter 9, in Neural Networks for Hydrological Modelling, Balkema, Rotterdam.
- The next step is for Linda to write a project proposal based on this item and email it to current Masters students to see if there is any interest.
-
- Browsing
- UoL
- Communication and Information Technology Strategy
- ISS faculty consultation to:
- Identify new services or changes to existing services (over the next 5 years) to improve:
- research
- learning & teaching
- education & knowledge transfer
- Address the following questions:
- What are our (top) long term academic objectives?
- What can ISS do to help achieve these goals?
- What should be provided by local IT and what should be central?
- How can IT best impact on academic priorities?
- How can IT provision best align to the business needs in the faculty?
- What are the most important things ISS should be doing?
- A mind map linking the UoL strategy map key themes to IT options
- Feedback sent to Mike Crabtree
- I think:
- We should be more involved in the development of open standards and open source software for all our needs.
- How is Leeds involved in W3C, OASIS, OGC and other relevant organisations? - How do we get more involved?
- If we don't develop skills in developing the software we want and use, then we are preventing the development of nurturing roots!
- In terms of processing capabilities and contribution to national and international e-Infrastructure. We should look to:
- Increase the provision and use of processing capability.
- Embrace WRG, NGS. and EGEE initiatives.
- Configure all computer clusters so they are available as a grid computer.
- Safer and more retrievable archives are a good thing.
- The importance of openness and of detailing the make up and process of our organisation using structured information should not be underestimated.
- Geomorphometrics
- What is progress with MARS data processing?
- e-Science
- MoSeS
- Email communication with Richard Long about Health Data
- NeSS
- Telephone call with June Finch
- I briefed June about the SEE-GEO Meeting on 2008-03-20
- June gave me permission to charge expenses to the project :-)
- June pointed out that I had forgotten that the Grid Enabling of datasets Work Package 2 was also relevant and that I should have mentioned this and SARoNGS in my presentation.
- New date for a meeting in Leeds
- June still aware that I want sign off criteria for deliverables.
- Project meeting on 2008-03-13
- It went ahead and was interesting, but no documentation is available.
- I explained roughly what happened from my end.
- We expect that next time the meeting will work better so long as there are no network problems.
-
-
- Attended some of the UPTAP Workshop 2008
- Chat with Cecilia MacIntyre (GROS), Paul Boyle and Mathew Hughes(?) at lunch:
- John Stillwell introduced us and we talked first about blogging :-)
- Is everyone in your group blogging?
- No, but every researcher should be :-)
- I first met Cecilia at an UPTAP Workshop in 2006
- http://www.geog.leeds.ac.uk/people/a.turner/personal/blog/archive/2006/03/#2006-03-31
- Following this Cecilia's colleague Ganka Mueller sent me some links to work that I hoped to feed back on which I still think we've yet to get round to:
- I still plan to feedback on Raj Bhopal's report...
- Laterly, I emailed Mark Birkin and Justin Keen to find out if MoSeS had a take on this that was any different to that stated in the emails of yesteryear.
- We mentioned the work of Pablo Mateo and the names projects at UCL
- Paul recently examined Pablo's thesis :-)
- Paul says that names and addresses were digitially stored for the 2001 Census and that there has been work done linking this with medical records to attach ethnicity information.
- This is very interesting and I'd like to learn more about it.
- I wondered if any work had been done with the UK cancer registry...
- Paul says that in 1991 there was a statement on the front page of the census form that names would not be digitally stored.
- There are problems with the 2001 Census name encoding:
- There were not separate boxes for forename and surname, but a single box for name, so all number of combinations are in there.
- e-Science
- GENESIS
- Environment and Planning B: Planning and Design 2008, volume 35, pages 191 -194: Editorial by Mike Batty
- I like the editorial :-)
- It is good to get some history and think about if parsimony and the generation of hypothesis is relevant.
- I prefer: so-called "black box" models that represent well, but which are understood less, than; simple models, which although intuitively developed and often declared to be 'more easily understood' - represent less well.
- I understand how artificial neural networks work and appreciate their capabilities. I do not need to know necessarily which factors combine to have confidence in their outputs.
- Models suited to representing complex systems are necessarily complex.
- The decomposition of a complex system into separate components (probably also complex systems) and the mapping of linkages between them (topology) is on-going science.
- I predict that the information systems (organisations) of the future will developed fast and in a largely automated way.
- I don't think "The Simulation" that Andy Evans mentions is just around the corner, but I do think there will be a time when we mortals will have to leave it to the system.
- I could be stating the obvious, but in order to map out the system we focus a window, set our scales and operate synoptically at first, but as with all automation, we finally let go.
- In terms of language, Mike has used the term 'computable' in "computable devices" where I originally thought 'computational' or 'computation' was more appropriate.
- In the context in the editorial I think the semantic of 'computation' was intended and what was arrived at (possibly unintentionally) is more interesting:
- What devices are computable?
- In science, parsimony is preference for the least complex explanation for an observation.
- Parsimony in theory goes hand in hand with the generation of hypotheis.
- Is it likely that the formation of hypotheis is irrelevant?
- Well, that all depends...
- Anyway, we need a map to start even if this is all we really need ;-)
- We need information to generate knowledge.
- We need a syntax for semantics.
- Ontological social science, generative social science, same difference?
- References
-
- Browsing
- Attending some of the UPTAP Workshop 2008
- NCeSS Sakai Portal : My Workspace Wiki : 2008-03-18
- Met Daniel Guinea-Martin and chatted about MoSeS :-)
- Attended the morning session with an interesting presentation from Paul Norman on Demographic and deprivation change in the UK 1991-2001:
- Paul has converted 1991 ED data to 2001 OA boundaries using address densities :-)
- I am interested to learn more about this and weather it can be part of ConvertGrid...
- I mentioned density type scatterplots to Paul and he seemed interested :-)
- References
-
-
- Browsing
- Provided feedback to Nanlin Jin on a paper titled "Genetic Algorithms for Dynamic Land-use Opimization" which has been accepted for presentation at the World Congress on Computational Intelligence (WCCI 2008)
- UoL/CCG/SoG/C&B collaboration opportunity
- How Ideas Travel (HIT) project
- Meeting with Jo Sadgrove
- Jo looking to get networked with expertise on mapping/visualising WWW content.
- I think Jo should talk with the SECSE people here in geography and grow links with our colleagues in the School of Computing.
- I am interested in this work.
- I really want Jo to get her home page sorted out :-) (At the time of writing the link above does not reach her home page.)
- I did a bit of a scan for some links to information, but did not find what I wanted, here is what I browsed:
- Geomorphometrics Meeting with Steve Carver and Justin Washtell
- They want to produce some visibility results for a Meeting on Monday and wanted advice on how to do this.
- There are 24 separate jobs that need to be run on a windows machines with around 4GB of RAM.
- NGS options are not in scope in this time frame.
- The best option is probably to make use of the Masters computing lab machines and have a way of it being easy for School of Geography colleagues to run jobs on their PCs over the weekend (a scavenger type grid).
- e-Science
- MoSeS
- Meeting with Junaid to look at Paul's EDINA GLS Client Java source code:
- Copied Paul's code from /home/tomcat/gridsphere-2.2.15/projects/edinageolink/ on geo-s12.leeds.ac.uk to my PC and set up a NetBeans project.
- Fetched /home/tomcat/gridsphere-2.2.15/src/javax/ and added this to the src directory and were able to compile ExecuteEDINAGeolink.java.
- Specified the javax.portlet imports and did some random documentation improvements.
- Continue to assume Paul will be happy to release his code under an LGPL license so that we can improve and enhance it.
- I bundled up the edinageolink directory on my PC into a zip file and uploaded this to the NCeSS portal. It should be available via the following URL to Junaid:
-
-
- Browsing
- SoG
- Feedback on Issue 90 of the SoG Newletter item "What do you Think?"
- Here is the item:
- Recently its been noticed that on a few occasions a senior colleague has presented the work of another, usually junior and/or admin/support colleague as their own, without giving due credit to the originator of the work. It may have been in a presentation, a paper, a report etc.
- Subsequently, the senior staff is credited with the work and due thanks etc are offered to them.
- To them, it may seem ‘no big deal’ to take someone else’s ideas, proposals, data gathering etc and to use them in their final piece of work, but for the person who has contributed - their contribution should at least be acknowledged – either with their name added to the paper or acknowledged at the presentation etc.
- There are some of the University values that seem to be being ignored in this practice:
- Where is this demonstrating:
- Would other colleagues agree that we all should give appropriate credit to collaborators within the School – even when it’s just for School or University papers, reports, presentations etc?
- No-one should take the credit for someone else’s contribution.
- What do others think?
- I sent the following by email to David Appleyard and Angie Grain:
- IMHO:
- Anyone who contributes to the production of some work should be acknowledged in a publication of it. Only those that make a substantive contribution to a publication should be in a list of authors. As with everything, there is a grey area here. I have seen data usage agreements which in the license make it a requirement for the data use that certain data providers are listed as co-authors on the publication of research that uses it. Such data licensing is perverse. Where this is the case, it should be made clear in the publication or in some reference or at least in some other publication that references the publication. It is probably easier to expose such perversions and make good use of the data others provide than to disregard such data on matter of principle.
- It should not be assumed that someone wants to be listed as an author if they have not made a substantive contribution to the publication. My name has appeared on publications without me putting it there or making a substantive contribution.
- Furthermore, an author list should not be assumed to be an endorsement of the conclusions of a publication unless it is single authored or clearly stated. Indeed, if there is disagreement it should be clear that there was such disagreement and that the article is an attempt to represent consensus, or that comclusion X is drawn by a specific set of authors...
- I hope this is a worthwhile contribution to this thread.
- CCG
- Xiaofu Guo enquiry about RLF/1
- Xiaofu graduated from Leeds in 2006 in Environmental Management.
- I plan to meet with Linda See about scoping out documenting and getting this working again.
- No date set, but hopefully next time Linda is in Leeds in around two weeks time.
- e-Science
- MoSeS
- MoSeS meets NECTISE part 2
- Meeting with Mark
- This was an impromtu meeting.
- Documentation
- In general the results were encouraging, but little analysis could be performed as more information was needed.
- We agreed that Mark would improve the table by:
- adding correlation statistics columns for all 3 comparisons.
- produce the table for a larger number of variables (say 50).
- Further discussion:
- Covariate analysis with the SARs:
- Like Dan Vickers did at the area level.
- Factor analysis.
- How rare is an individual and the representativeness capability of the SARs.
- Boosting the HSAR.
- The ISAR only results are now more interesting to Mark.
- Meeting with Junaid
-
- Browsing
- GEOG5060
- Preparation for practical:
- Assuming all PCs in the GIS lab have a C:\temp directory that can be written to by all users.
- e-Science
- e-Uptake Meeting/Interview
-
- Browsing
- e-Science
- Community Engagement and Support Projects, JISC e-Infrastructure Programme
-
- Browsing
- e-Science
- EDINA arranged SEE-GEO telecon
- Chris Higgins minutes emailed same day :-)
- MoSeS
-
- Geog5060
- Delivered the lecture and gave a demo/workshop.
- It seemed to go well, infact better than last year:
- I have last years students to thank in part for this :-)
- There are at least four Java programmers in the group :-)
- Unfortunately there are no multidisciplinary informatics students this time.
- Andy Turner's GEOG5060 2008 Web Page
- Meetings with Oliver Duke-Williams
- Web GIS
- Oli teaching a course in this.
- A new or improved School of Geography web server would be useful.
- CCG Web pages
- Andy to add links to:
- Oli's teaching
- UPTAP, ONS in groups
- Software Web Page something for René to develop?
- Desire for a data page
- Demo of MoSeS GeoServer and Gridsphere Portal
- Blogging and some recent meetings:
- CSAP meeting with Crispin Cooper
- We looked at 1991 and 2001 migration data for London as pulled from WICID and saw no obvious problem with Lambeth.
- Oli will email Crispin
- Today's CSAP meeting
- Summarised Chengchao's work.
- Chatted about René's work.
- Data
- OpenStreetMap OpenAerialMap GoogleMaps and OpenSpace
- References
- There is a big difference between a disclosure and the ability to make a disclosure:
- As census users we operate in a legal framework and we sign forms to use the data under the terms and conditions:
- Each user is responsible for ensuring derived data is non-disclosive, or disclosive in ways which are agreed to be allowed.
- When the ability to make a disclosure is constrained you can quickly end up with a census that is obfuscated and hard to use like the 2001 UK Human Population Census.
- There is a massive waste of resource.
- Much effort has been focussed on trying to process this census data to produce data that could have been made available directly.
- I fell like a victim of this GIS crime!
- XML Census 2001, 2011 and MoSeS:
- I described the preprocessing and data access of the Census Aggregate Statistics we are using in MoSeS.
- This would have been much easier if the metadata and data were in an XML form.
- XML is being proposed as the de facto data format for the 2011 UK Human Population Census:
- If they get it right, we will even be able to request specific versions of the data.
- http://www.geog.leeds.ac.uk/people/a.turner/projects/MoSeS/documentation/demography/results/tests/sourceData/CAS001vCAS002AgeGroupComparison/CAS001vCAS002AgeGroupComparison.xhtml2.0.html
- Are some of these points so far from the line because I have loaded the data wrong for one country?
- The points should be pretty much on the line...
- Anyway, it should be straight forward to plot two variables that should in theory represent the same counts from different aggregate statistics tables like this:
- It is a way to check for errors and data consistency and to have some idea of the sensitivity of different variables and tables to disclosure control measures.
- e-Science
- NeSS Project Meeting with June Finch
- MoSeS
- Community Engagement and Support Projects, JISC e-Infrastructure Programme
-
- Browsing
- http://geothought.blogspot.com/2008/02/if-you-could-do-geospatial-analysis-50.html
- Added comment:
- Hi, I've been working on Geoprocessing a bit and am developing some software for this. One of the components is a Grids Library.
- OSGeo
- GEOG5060 Preparation
- Steve suggested the use of GIS Tutor software in the practical on Friday and to talk with Helen about this:
- The software is "Net Support School" allows for demonstrations in the GIS Lab whereby all screens show the screen output on a specific PC.
- Helen gave me login credentials and basic instructions :-)
- Geomorphometrics
- Progress on NextMap Cairngorm run:
- Iteration 1: Done
- Iteration 2: java.lang.StackOverflowError
:
- e-Science
- NEC MoSeS Demo Preparation
- Applied for renewal of UK e-Science Certificate.
- Meeting with Junaid Arshad
![[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)