際際滷

際際滷Share a Scribd company logo
Effie Kapsalis, Smithsonian @digitaleffie
Imagine the Future of Righting Our
Wrongs
IGNITE MCN
NOVEMBER 5, 2019
When youre invisible,
people assume youve done
nothing.
Edith P. Mayo, curator emerita, Smithsonian's National Museum of American
History
MCN 201902
Whose history is this?
MCN 20193
Where are the Women? A Report on the Status of Women
in the U.S. Social Studies Standards
http://bit.ly/349OmrL
MCN 201904
Where are the Women? A Report on the Status of Women
in the U.S. Social Studies Standards
http://bit.ly/349OmrL
MCN 20195
18% of Wikipedia bios are
about women*
*It took Wiki Women in Red 5 years working internationally to
increase this number from 15.53% to 18.1% in September 2019.
MCN 20196
MCN 20197
No need for text, so delete me...
MCN 201908
MCN 20199
MCN 2019010
Improved the safety of a multi-million dollar
industry. A single line in a finding aid.
MCN 2019011
MCN 2019012
African American advertising
The original Lois Lane, founder of the Black
Fashion Museum, 1979. Deemed not notable.
MCN 2019013
CrowdsourcingDigital Curators Machine
Learning
New Digital Resources on American Womens History
Smithsonian American Womens History Initiative
Collections as Data
Digital
Curators
Surface hidden figures to
create structured data and
trusted digital resources.
MCN 201914
Caroline Jones
Spearheaded
advertising
campaigns to include
diverse populations
and changed the way
Americans thought
about popular brands.
The Funk List: Guerilla History
MCN 201915
MCN 201916
Machine
Learning
Analyze representation,
improve data, and share
results at-scale.
Zitk叩la-叩
Sioux activist
who worked for
the reform of
United States
Indian policies.
Visualizing the landscape of
digital data with topic modeling
MCN 201917
"topic":["Flowers","Equipment","Needlework","Textile
Working","Trees","Women","Sewing","Architecture","Nature","Ecology","Porch","Gardens]
"Plants","Portraits]
"topic:["Women","Society and social change","Portraits","Hair
Accessory","Landscapes","Necklace","Costume","Socialite","Jewelry","Pearl]
"topic:["Women","Big band music","Jazz","Musicians","Bands (Music)]
:"topic:["Dance","Figure group","Adobe","Rites and
ceremonies","Women","Domestic","Architecture","Landscapes]
MCN 201918
Crowdsourcing
Invite the crowd in to
surface stories, improve
data, and share results.
Sandra Cisneros
First Mexican-
American writer to
have her work
published by a
mainstream American
publisher.
Words matter
MCN 2019019
#BecauseOfHerStory
MCN 201920
THANKS!
Effie Kapsalis, Senior Digital Program
Officer Smithsonian American Womens
History InitiativeIGNITE MCN
NOVEMBER 5, 2019

More Related Content

Imagine the Future of Righting Our Wrongs

  • 1. Effie Kapsalis, Smithsonian @digitaleffie Imagine the Future of Righting Our Wrongs IGNITE MCN NOVEMBER 5, 2019
  • 2. When youre invisible, people assume youve done nothing. Edith P. Mayo, curator emerita, Smithsonian's National Museum of American History MCN 201902
  • 3. Whose history is this? MCN 20193 Where are the Women? A Report on the Status of Women in the U.S. Social Studies Standards http://bit.ly/349OmrL
  • 4. MCN 201904 Where are the Women? A Report on the Status of Women in the U.S. Social Studies Standards http://bit.ly/349OmrL
  • 5. MCN 20195 18% of Wikipedia bios are about women* *It took Wiki Women in Red 5 years working internationally to increase this number from 15.53% to 18.1% in September 2019.
  • 8. No need for text, so delete me... MCN 201908
  • 10. MCN 2019010 Improved the safety of a multi-million dollar industry. A single line in a finding aid.
  • 12. MCN 2019012 African American advertising The original Lois Lane, founder of the Black Fashion Museum, 1979. Deemed not notable.
  • 13. MCN 2019013 CrowdsourcingDigital Curators Machine Learning New Digital Resources on American Womens History Smithsonian American Womens History Initiative Collections as Data
  • 14. Digital Curators Surface hidden figures to create structured data and trusted digital resources. MCN 201914 Caroline Jones Spearheaded advertising campaigns to include diverse populations and changed the way Americans thought about popular brands.
  • 15. The Funk List: Guerilla History MCN 201915
  • 16. MCN 201916 Machine Learning Analyze representation, improve data, and share results at-scale. Zitk叩la-叩 Sioux activist who worked for the reform of United States Indian policies.
  • 17. Visualizing the landscape of digital data with topic modeling MCN 201917 "topic":["Flowers","Equipment","Needlework","Textile Working","Trees","Women","Sewing","Architecture","Nature","Ecology","Porch","Gardens] "Plants","Portraits] "topic:["Women","Society and social change","Portraits","Hair Accessory","Landscapes","Necklace","Costume","Socialite","Jewelry","Pearl] "topic:["Women","Big band music","Jazz","Musicians","Bands (Music)] :"topic:["Dance","Figure group","Adobe","Rites and ceremonies","Women","Domestic","Architecture","Landscapes]
  • 18. MCN 201918 Crowdsourcing Invite the crowd in to surface stories, improve data, and share results. Sandra Cisneros First Mexican- American writer to have her work published by a mainstream American publisher.
  • 21. THANKS! Effie Kapsalis, Senior Digital Program Officer Smithsonian American Womens History InitiativeIGNITE MCN NOVEMBER 5, 2019

Editor's Notes

  • #3: Representation matters People need to see themselves in history For cultural orgs, its a matter of relevance. Get your phones ready for the last slide.
  • #4: Revise standard K-12 American history timeline As long as history follows the traditional timeline, the experiences of women and diverse U.S. populations are marginalized WE CAN DO BETTER
  • #5: The barriers to entry into K12 history curriculum are high. Racial distribution of the 178 women 63% White, Zero Asian American women? disability rights, Native American rights, LGBTQ, or migrant labor movements are way underrepresented WE CAN DO BETTER
  • #6: The biases online reflect our culture Took 5 years for international Wiki women in red project to improve representation by just under 3% Same biases in our collections. A giant ingest of Japanese footballers can bring this number back down. WE CAN DO BETTER
  • #7: Look inside. Where is the context? The stories? Barrier for entry is too high WE CAN DO BETTER
  • #8: Google does it. Books, context, images. Thank you Wikipedia. Why dont we go the way of Tate and add Wikipedia to our websites? WE CAN DO BETTER
  • #9: Digital trail of one woman in Smithsonian Roxie Collie Laybourne was single line in finding aid. Anyone able to tell why shes important? A plan took off in 1960 from Boston Logons airport plane crashed 6 seconds later and 62 people died. Sent feathers to Smithsonian?
  • #10: With the majority of staff blogging, we start to create context w/ blog posts and WISW Secondary sources
  • #11: Started new field of science and helped solve criminal cases. Taught taxidermy to city kids. Do your work and keep your mouth shut, and they will respect you, man or woman. And thats why I got along as well as I did in a mans world.
  • #12: After an edit-a-thon in 2012, she averages 1000 views/month on Wikipedia Have we taken this bio back? I CAN DO BETTER
  • #13: Wikipedia has problems. A wiki newbie wrote this article Marked for take down Women and diverse populations have less secondary souces Hey, WE CAN DO BETTER
  • #14: We gotta do it at scale As a National Museum complex, its on Smithsonian to test out
  • #15: Curators to comb collections for names and stories Instead of exhibits, structured data + digital resources
  • #16: AWHI Digital Curator, History of Women in Science Building on staff crowd-sourced list Named in honor of Dr. Vicki Funk Reached out to her network for female firsts and seconds, even if didnt have scientist title 125 names in structured data Now over 400
  • #17: Can machine learning help w/ gender representation when tools have biases. Dunno But it can help analyze patterns & improve things at scale
  • #18: Smithsonian Data Science Lab Collection recordsonline , only 100,000 of 14 million are tagged with women in the topic fields. Topic fields for records tagged with women. Natural Language Processing and Topic Modeling to analyze what topics more likely to be associated with women than others? Information for curators
  • #19: Seen crowds surface women They share our resources more than we ever can Lets diversify what they do and involve the world building middle school curriculum for Smithsonian TC to transcribe girl history
  • #20: Im sorry but our vocabularies could use a major update. We dont come close to the nuances of gender. WE CAN DO BETTER Phones ready?
  • #21: Snap this and add to our worklists with your resources These women have been sidelined and silenced long enough. Taylor Swift Im so sick of running as fast as I can wondering if Id get there quicker if I was a man.