Reading and Linking Place: Text on Historical Maps @UCGIS Symposium 2022

JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.

Thank you for your feedback, we truly value your contribution. Your opinions and suggestions will help us shape the future development of the tools you have just tested, so that they meet more effectively the needs of our different users.

The data in this form is anonymous, and will not be shared with anyone outside this project. Brief, anonymous quotes from these feedback forms may be referenced in future publications. Feel free to answer only part of the questions and to submit an incomplete form. The workshop is sponsored in part by the AHRC/NEH grant (Award #HC-278125-21) to the Turing Institute, the University of Southern California, and the University of Minnesota. The tools developed under this project are open-source software and available online to the general public.

SECTION ONE: About Yourself.

We would like to know something about your background. Feel free to select as many options as appropriate.

Please, tell us more about your interest in annotating historical maps. Are you: *

a data curator/data manager (e.g., librarian) who needs to create metadata for historical maps

a researcher who wants to efficiently analyze and compare the full text content from one or multiple historical maps

a researcher who wants to build machine learning models to automatically extract information from historical maps (e.g., text detection)

Other:

Required

If you work with historical maps, how do you find and access maps? *

using university libraries (with help from librarians)

using university libraries (self-help)

using online map archives

using government agencies

Other:

Would you consider yourself familiar with GIS technologies, such as QGIS or ArcGIS? *

I work with GIS technologies regularly, I am a proficient user

I am familiar with the basics of GIS technologies, and I have used the software occasionally

I am not familiar with GIS technologies.

Other:

SECTION TWO: About the automated process

We shared with you two maps: one representing high false positive results with complex background and the other one representing relatively accurate results (i.e., Historical map of California, Map of Dublin 1847)

Have the automatically detected text bounding boxes saved you time from your annotation tasks? *

minimum save, I can live without it

great time saving

What could be the challenges of adopting mapKurator to your maps? *

my maps of interest are similar to the map of Dublin (or even more complex) and I expect there could be a lot of false positives

my organization does not have the resource (e.g., IT expertise / computer resource)

Other:

What is the main reason that prevents the automatic process to be useful? *

Missed too many text labels

Had too many incorrect labels

Other:

How relevant is for you to be able to generate a list of possible semantic types (e.g., city names and road names) for a text labels

not very relevant

extremely relevant

Clear selection

SECTION THREE: About the communities

From your experience, what would be the communities that will be interested in using the tool or collaborating in the development efforts? What would be the places to meet these potential communities (e.g., certain conferences, workshops) *

SECTION FOUR: Share your annotation results

We will only use the data anonymously for the purpose of improving mapKurator and Recogito. We will not share the data.

Please share your link to the annotation results *

Submit

Clear form

Never submit passwords through Google Forms.

This form was created inside of University of Minnesota Twin Cities. Report Abuse

Forms