Human-Computer Collaboration at NYPL Labs

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.

7aff8f547184534da3ca2e14e63a68a8?s=128

Mauricio Giraldo

April 14, 2016
Tweet

Transcript

  1. None
  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

  6. None
  7. None
  8. Eric Shows

  9. NYPL Labs

  10. access digitization metadata public traditional digital library program

  11. access digitization metadata public engagement r+d

  12. None
  13. None
  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

  29. None
  30. None
  31. None
  32. None
  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
  35. None
  36. None
  37. None
  38. *this is a simulation. actual process is intensive. consult your

    mathematician before trying
  39. None
  40. None
  41. and now you start tracing those buildings by hand (˽°□°)˽Ɨ

    ˍʓʓˍ
  42. None
  43. 1852-1854

  44. 1852-1854

  45. can we automate this?

  46. computers are good at some things…

  47. None
  48. None
  49. None
  50. None
  51. None
  52. None
  53. None
  54. yay footprints! 60,000+ of those!

  55. like OCR for maps!™ (not really trademarked)

  56. but OCR is pretty bad ಠ_ಠ

  57. None
  58. people are good at other things!

  59. None
  60. None
  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

  65. None
  66. what about malicious users? or even well-meaning ones who make

    mistakes
  67. None
  68. 75% or more agreement between 3 or more people arbitrary

    numbers that have worked for us
  69. None
  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?

  73. None
  74. None
  75. None
  76. footprint material use type street names address floors name class

    geo location year skylights backyards
  77. None
  78. None
  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

  84. None
  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)

  88. None
  89. None
  90. None
  91. good tutorials are hard

  92. None
  93. Super Mario Bros. (Nintendo, 1985)

  94. we have too many edge cases or: how i learned

    to stop worrying and embrace imperfection
  95. None
  96. ¯\_(ϑ)_/¯ people skip them anyway

  97. None
  98. coming soon

  99. None
  100. None
  101. None
  102. None
  103. NYPL Community Oral History Project oralhistory.nypl.org

  104. None
  105. None
  106. make these stories more accessible

  107. None
  108. mark transcribe

  109. by brian foo @beefoo

  110. allows for basic text search but it’s not a proper

    transcript
  111. None
  112. None
  113. we felt we needed something different

  114. None
  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
  118. None
  119. people are good at other things…

  120. None
  121. None
  122. by brian foo @beefoo

  123. we conducted a few usability studies

  124. by brian foo @beefoo

  125. None
  126. None
  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…

  139. None
  140. None
  141. None
  142. None
  143. lala.cursivebuildings.com

  144. how to decode the 3D data? in the browser

  145. None
  146. None
  147. None
  148. None
  149. None
  150. stereo.nypl.org

  151. None
  152. None
  153. None
  154. None
  155. None
  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