Notes: Geo Tools, Geo Parsing & GIS (+ Collaboration)

November 20th, 2010 |

I took these notes on the GeoTools/GIS session, which was later joined by a group from the Collaboration session.  Feel free to edit.  These are not exhaustive at all and are unedited.  Feel free to modify/correct/add.


10 am – First Session: Geo Tools, Geo Parsing & GIS – RM 2370 – B


Samip Mallick –  U of C
-Applied for a grant for interface for census data. Like social explorer.

Tobias Boes – Notre Dame German instructor
-New things spatial mapping can tell us (literature) how does it change the intuitive lessons of the novel.
-Interested in Hyper Cities – Historical layers of maps, add annotations.
-How does this apply to language courses.

Kirk Hess
-Text encoding. Chemical writings of Issac Newton, transferred into TEI. Multimedia cataloging/archiving.
-Prob: locations are often named by the person who made the annotations/materials.  Place names change over time. Cities grow and shrink.
-Mapping videos of Vietnam War. City of Memory
-City Stories – NYC – Uses map to show where stories are from.
-Issues with things like GoogleMaps – not open source.  There are alternatives. Postal service sold zip codes to private companies.  Geography becomes proprietary. Open Source: Geo Server – akin to Google Maps.
-Sometimes Google is the easiest thought, – struggle

Ron Zacharski
-Lesser know languages
Interested in adding visualization component.

Gemma Petrie
-New to DH.
-Worked as nonprofit development director.  Worked with data that was hard to use. Implications of GIS and visualization are huge for both research and policy and fundraising.

Christopher Dickman
-Working on digital multimedia compendium at St. Louis University with a government grant to redo how they do their first year composition classes. Cognitive writing theory.
-How can we teach with this.
-Data and map visualization of authors lives.

Marjorie McLellen
-Personal narratives, place and landscape, people’s memories of their childhood and mapping that.
-Dayton flood – mapping

Historian at U of C
-19th century intellectual history
-Interested in geo parsing – methodological histories of landscapes.
-Novels- both real and imaginary places.  Map fictional places automatically by mapping real places in text.
-Giving a talk tomorrow

Jo’s work
-Sounds like a named entity task (Ron). If you have already marked up some text, you can run a dining mining algorithm to teach the computer to learn to extract these names.
-Kirk – will need to employ someone to clean up data. Is there an algorithm for this already?
-Ron – if you have a term, analyze words around the text to determine if the term refers to two different entities.
-Ron and Jo might collaborate.
-Ron- uses Python
-Christopher – DocuScope. – Will parse text for words about reason – jokes, – arguments, positive and negative feelings. (Recently in Shakespeare Quarterly).  Did a great job determine which plays were comedies and tragedies, etc. Examples, they found a tradgedy with lots of comedic language.
Ron- Do you have to train DocuScope?
Christopher- Created by Carnegie Mellon. One person, so somewhat personal.  Subjective in terms of what means positive, negative, etc.  Could be good for searching for things like place prepositions.
-Jo – once you have pulled the terms referring to Bohemia, you can analyze text around to see if it is a happy place, etc and if that changes over time.
-Kirk – Benefits of GeoServer – you can make up maps, not constrained by the Earth like GoogleMaps.
-Jo – Geo Rectifying – NYPL does a lot of these projects (taking maps and putting it on top of data). Crowdsourcing information
-Samip – MapWarper: – you can use this application and upload your own data. Once you have the geo rectified map, you can place it on to Google Maps or Geo Server

Samip: Stanford – Mapping the Republic of Letters.
Gemma: I think this is the project that lead scholars to question if the Enlightenment originated in England
Gemma: Also Spatial History at Stanford:
Jo- Stanford setting high bar. But, they aren’t really tools – they are visualizations for their own research. Jo has tried to contact them to use their tools, with no response.  Makes her frustrated that this information isn’t shared. They use GIS and highly advance visualization talents.  Nothing new, simply the best geographers in the world right now. This is not collaborative.

Samip- There is still a huge learning curve to open source tools.

Kirk and Majorie: NYC
People can upload their own content
People nominating sites would guide the work of the developers.  They would go out and research sites.  Guided by users.
Tobias: More collaborative than Stanford
Marjorie: Also a neat Run Lola Run map.

(We were joined by a bunch of other Campers from the Collaboration group)

Samip: What would make something useful to scholars? Problem with Stanford, we are only able to see what they want us to see.
Jo: She sees an issue in her teaching.  She assigns tools that are almost always proprietary – come from the private sector and not meant to help the community as a whole, just to support a few dissertations.

Karl Stadely: Issues of developing talent. What are we doing to cultivate talent? Why does programmer always have to be some outside person? Pulling out one element shouldn’t change the nature of the whole.
Marjorie – must be mindful of the potential of our students.  Must attract students into the humanities that have these talents – not just to exploit them – but because they will find exciting work to be done.
Karl: Tech is such a talent everyone outside of the academy (teaches info arch at IIT)
Peter Organiscak: a central hub to foster collaboration. No place for the community to go search for tools before you create yourself. (Or, to find something to join in on). Often, we retread the same tracks in DH. We need a place of record.
Ryan: Collaborating on building a tool to scrape newspapers for spatial history. Alternatively, it would be interesting to build something that would grow.
Peter: grad thesis: motivators of users, crowds and crowdsourcing: passion. People need to be able to find out about projects. If people could find it, students could collaborate too.
Christopher: might be issues of motivations/tenure/publications
Ryan: a central hub for DH knowledge. (DH Answers):

Jo- How do you keep people interested?
Ryan/Peter: need central management, invested parties
Karl: Maintain excitement by maintaining progress.
Jean Bauer: Need to be transparent about you got there (when building things) in open source projects.
Christopher: DH projects rather than DH answers. It would be good to be able to see what else is happening as you plan projects.
(attempt to find a URL to buy)
Peter: it would be helpful to know about project overlap before conference presentations, etc.
Ryan: Issues of credit in this kind of collaboration.
Jean Bauer: Clear acknowledgement
Kirk: Teaching Git – /git/blame Share intellectual work and credit.
Christopher: So, should we create something
Kyle R: What about? Center Net: Directory of DH labs:

Peter: Primary motivators (based on his thesis) of users: Passion/topic interest, ease of entry and participation, altruism/meaningful contribution (impact on the tool), sincerity (ownership/collaboration), money.
Ryan: No place to post your own project and find collaborators.
Kirk: We could just create a Git Hub account right now. Can be as permissive or restrictive as you want to be.
Sherman: Dork Shorts: 30 sec shorts.
Ryan: a way to say, Here’s an idea I have, anyone want to jump in?
Marjorie: Drupal Commons (like Buddy Press)
Issue: Git Hub vs. a nicer presentation with video shorts, etc.
Christopher/Marjorie: something like TED?
Christopher: A list of projects, tagged, easy to browse and search. Anyone can come to post their project. Maybe completed project and a active project sections. Project could be the tool or the result of the tool.
Ryan: A place to go when you have an idea to see if anyone else is doing something similar and if you should start something knew or try to get involved with something that is already happening.
Jo: Tibetan Himalayan Library: A collaboration/agreement of scholars working in this area to encode their notes/work in a particular way that allows this information to be available to others.
-Harvard GIS Project- You can see who else is interested in work on a certain area.  Perhaps the biggest thing out there for GIS collaboration in the humanites.
Jean: Stanford projects, looking for more data. Have worked with Electronic Enlightenment, now working with smaller projects. They use a simple MySQL Entities to Links database structure.

Lunch: Some members are going to continue planning a collaborative web-environment for projects.

Comments Feed

Comments are closed.