Human-Computer Collaboration at NYPL Labs

This talk is part of the Culture Analytics and User Experience Design Workshop organized by the UCLA's Institute for Pure & Applied Mathematics, April 11 - 15, 2016.

Abstract:
This talk will provide an insight into the Building Inspector, Oral History, and other projects developed at New York Public Library Labs, with an emphasis on design and HCI-related challenges and iteration processes. It will cover the rationale as to when NYPL Labs uses (or does not use) computational tools for data extraction and improvement, and how the library is learning to collaborate with its users on the creation of digital resources.

Transcript

2. human-computer collaboration mauricio giraldo arteaga @mgiraldo @nypl_labs IPAM Culture Analytics and User Experience Design, April 2016
3. hello
4. not a real library scientist
5. flickr.com/photos/wallyg/6133216510
8. Eric Shows
9. NYPL Labs
10. access digitization metadata public traditional digital library program
11. access digitization metadata public engagement r+d
14. what happens after digitization?
15. human-computer collaboration
16. ¿ ?
17. embrace imperfection corollary of “perfect is the enemy of good”
18. « A designer’s definition of ‘perfect’ is different for computational designers. » because it is not achievable John Maeda
19. human-computer collaboration
20. computers are good at some things…
21. Randall Munroe - xkcd.com/1140
22. David Hagen - drhagen.com/blog/the-missing-11th-of-the-month
23. people overestimate OCR quality
24. OCR result
25. okay… so maybe computers are not that good
26. people are good at other things
27. human-computer collaboration i avoid the term “crowdsourcing”
28. two examples
33. footprint material use type street names address floors name class geo location year skylights backyards
34. like Google Maps for the 19th century but Google Maps cannot answer questions about the 19th century
38. *this is a simulation. actual process is intensive. consult your mathematician before trying
41. and now you start tracing those buildings by hand (˽°□°)˽Ɨ ˍʓʓˍ
43. 1852-1854
44. 1852-1854
45. can we automate this?
46. computers are good at some things…
54. yay footprints! 60,000+ of those!
55. like OCR for maps!™ (not really trademarked)
56. but OCR is pretty bad ಠ_ಠ
58. people are good at other things!
61. people don’t choose to complete these
62. we have over 60,000 footprints to check! will people want to do this?
63. what is the minimum contribution we need? we want the lowest friction possible so people will want to contribute
64. this was 2013, touch-screen mobile had taken off
66. what about malicious users? or even well-meaning ones who make mistakes
68. 75% or more agreement between 3 or more people arbitrary numbers that have worked for us
70. YES is on the right side because most people are right-handed and the algorithm is right most of the time
71. Building Inspector buildinginspector.nypl.org
72. will people participate? remember that little tweet button?
76. footprint material use type street names address floors name class geo location year skylights backyards
79. check YES FIX address color fix *footprints marked as “NO” go to polygon heaven
80. address had to use full keyboard on mobile because fractions
81. classify
82. fix
83. place names
85. we add new maps as old ones are completed the bottleneck now became geo-rectifying those maps ¯\_(ϑ)_/¯
86. this is actually version 2
87. (the magic of git)
91. good tutorials are hard
93. Super Mario Bros. (Nintendo, 1985)
94. we have too many edge cases or: how i learned to stop worrying and embrace imperfection
96. ¯\_(ϑ)_/¯ people skip them anyway
98. coming soon
103. NYPL Community Oral History Project oralhistory.nypl.org
106. make these stories more accessible
108. mark transcribe
109. by brian foo @beefoo
110. allows for basic text search but it’s not a proper transcript
113. we felt we needed something different
115. computers are good at some things…
116. like OCR for audio!™ (not sure if they trademarked that)
117. we get transcription “snippets” from 1 to about 6 seconds long in varying levels of quality
119. people are good at other things…
122. by brian foo @beefoo
123. we conducted a few usability studies
124. by brian foo @beefoo
127. it’s hard to reach consensus ಠ_ಠ
128. embrace imperfection
129. transcribe.oralhistory.nypl.org
130. transcribe.oralhistory.nypl.org
131. made with customizability in mind
132. storyscribe.themoth.org
133. this is one week after launch
134. it is still being improved
135. two of several projects we’ve worked on so far
136. of human-computer collaboration
137. it’s a collaborative process Willa Armstrong, Shawn Averkamp, Paul Beaudoin, Brian Foo, Josh Hadro, Elizabeth Hummer, Ara Kim, Shana Kimball, Tom Listanti, Matthew Miller, Eric Shows, Bert Spaan, and more at NYPL…
138. one more thing…
143. lala.cursivebuildings.com
144. how to decode the 3D data? in the browser
150. stereo.nypl.org
156. Boston Public Library Boston Public Library U.S. Geological Survey U.S. Geological Survey
157. thank you! mauricio giraldo arteaga @mgiraldo @nypl_labs IPAM Culture Analytics and User Experience Design, April 2016