Tweet
Tweet

Slide 1

Slide 1 text

Mauricio Giraldo Arteaga - July 2020 An experimental bird’s eye view of the digital collections from the State Library of NSW

Slide 2

Slide 2 text

hello

Slide 3

Slide 3 text

No content

Slide 4

Slide 4 text

Universidad de los Andes 2000

Slide 5

Slide 5 text

No content

Slide 6

Slide 6 text

Carnegie Mellon University 2011

Slide 7

Slide 7 text

No content

Slide 8

Slide 8 text

No content

Slide 9

Slide 9 text

No content

Slide 10

Slide 10 text

No content

Slide 11

Slide 11 text

No content

Slide 12

Slide 12 text

i was there — Photo: Paula Bray

Slide 13

Slide 13 text

No content

Slide 14

Slide 14 text

No content

Slide 15

Slide 15 text

No content

Slide 16

Slide 16 text

No content

Slide 17

Slide 17 text

No content

Slide 18

Slide 18 text

No content

Slide 19

Slide 19 text

No content

Slide 20

Slide 20 text

No content

Slide 21

Slide 21 text

No content

Slide 22

Slide 22 text

No content

Slide 23

Slide 23 text

No content

Slide 24

Slide 24 text

collection.sl.nsw.gov.au/digital

Slide 25

Slide 25 text

about 2 million at the time of the fellowship about 4 million images

Slide 26

Slide 26 text

flickr.com/photos/prasenberg - Patrick Rasenberg

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

No content

Slide 29

Slide 29 text

No content

Slide 30

Slide 30 text

No content

Slide 31

Slide 31 text

bounded by the physical space serendipitous exploration

Slide 32

Slide 32 text

how does this work digitally?

Slide 33

Slide 33 text

No content

Slide 34

Slide 34 text

554,030 results

Slide 35

Slide 35 text

*not really possible due to technical limitations in pagination depth 13,000+ “pages”*

Slide 36

Slide 36 text

this is a common paradigm

Slide 37

Slide 37 text

“pagination”

Slide 38

Slide 38 text

No content

Slide 39

Slide 39 text

No content

Slide 40

Slide 40 text

No content

Slide 41

Slide 41 text

No content

Slide 42

Slide 42 text

13,000+ pages

Slide 43

Slide 43 text

you miss the forest for the trees

Slide 44

Slide 44 text

No content

Slide 45

Slide 45 text

pages.gseis.ucla.edu/faculty/bates/berrypicking.html

Slide 46

Slide 46 text

- Marcia J. Bates “the query is satisfied not by a single final retrieved set, but by a series of selections of individual references and bits of information at each stage of the ever-modifying search”

Slide 47

Slide 47 text

research as a serendipitous process

Slide 48

Slide 48 text

constrained by small set of results on view digital serendipity is limited

Slide 49

Slide 49 text

No content

Slide 50

Slide 50 text

No content

Slide 51

Slide 51 text

No content

Slide 52

Slide 52 text

No content

Slide 53

Slide 53 text

see the forest and the trees

Slide 54

Slide 54 text

No content

Slide 55

Slide 55 text

No content

Slide 56

Slide 56 text

“big picture” vs. detail

Slide 57

Slide 57 text

*that i could research, design, and implement in eight weeks this is my attempt at addressing this*

Slide 58

Slide 58 text

No content

Slide 59

Slide 59 text

“An experimental bird’s eye view of the digital collections from the State Library of NSW”

Slide 60

Slide 60 text

but first some background

Slide 61

Slide 61 text

No content

Slide 62

Slide 62 text

No content

Slide 63

Slide 63 text

dependent on text metadata

Slide 64

Slide 64 text

No content

Slide 65

Slide 65 text

No content

Slide 66

Slide 66 text

No content

Slide 67

Slide 67 text

No content

Slide 68

Slide 68 text

comprehensive metadata is hard

Slide 69

Slide 69 text

we could automate metadata creation

Slide 70

Slide 70 text

we could automate metadata creation som e* *the one that computers are good at

Slide 71

Slide 71 text

already being done at the Library

Slide 72

Slide 72 text

sl.nsw.gov.au/blogs/tiger-using-artificial-intelligence-discover-our-collections

Slide 73

Slide 73 text

collection.sl.nsw.gov.au

Slide 74

Slide 74 text

No content

Slide 75

Slide 75 text

No content

Slide 76

Slide 76 text

computers get (a lot) wrong

Slide 77

Slide 77 text

No content

Slide 78

Slide 78 text

No content

Slide 79

Slide 79 text

No content

Slide 80

Slide 80 text

*there are way more problematic tags (that reflect algorithm biases) take it with a grain of salt*

Slide 81

Slide 81 text

Microsoft COCO: Common Objects in Context

Slide 82

Slide 82 text

computer “learns” how to “see”

Slide 83

Slide 83 text

No content

Slide 84

Slide 84 text

= “dog”

Slide 85

Slide 85 text

an issue in a collection with centuries-old stuff cannot see… what it hasn’t seen before

Slide 86

Slide 86 text

*non-W.E.I.R.D. stuff (Western, Educated, Industrialized, Rich, “Democratic”) cannot see… what its creator didn’t train* it to see

Slide 87

Slide 87 text

useful for finding similar-looking things, not for the categories themselves but it is consistent

Slide 88

Slide 88 text

No content

Slide 89

Slide 89 text

No content

Slide 90

Slide 90 text

with computers, avoiding closed or proprietary algorithms a way to group images together

Slide 91

Slide 91 text

image-net.org

Slide 92

Slide 92 text

ml4a.github.io/guides - Gene Kogan

Slide 93

Slide 93 text

No content

Slide 94

Slide 94 text

No content

Slide 95

Slide 95 text

or, in math parlance, uniform manifold approximation and projection (umap) “similarity metadata”

Slide 96

Slide 96 text

dhlab.yale.edu/projects/pixplot

Slide 97

Slide 97 text

dhlab.yale.edu/projects/pixplot

Slide 98

Slide 98 text

No content

Slide 99

Slide 99 text

No content

Slide 100

Slide 100 text

SFMoMA Artscope - Stamen (2014)

Slide 101

Slide 101 text

bertspaan.nl/semia - Bert Spaan

Slide 102

Slide 102 text

amnh-sciviz.github.io/image-collection - Brian Foo

Slide 103

Slide 103 text

we can also use color as metadata

Slide 104

Slide 104 text

No content

Slide 105

Slide 105 text

mkweb.bcgsc.ca/colorsummarizer

Slide 106

Slide 106 text

No content

Slide 107

Slide 107 text

artsexperiments.withgoogle.com/artpalette

Slide 108

Slide 108 text

collection.cooperhewitt.org

Slide 109

Slide 109 text

publicdomain.nypl.org/pd-visualization - Brian Foo

Slide 110

Slide 110 text

No content

Slide 111

Slide 111 text

“I propose enhancing the Manuscripts, Oral History and Pictures Catalogue with computer-generated metadata to create new pathways for patron exploration. By focusing on the 300,000+ digitised image set that is currently available…”

Slide 112

Slide 112 text

“I propose enhancing the Manuscripts, Oral History and Pictures Catalogue with computer-generated metadata to create new pathways for patron exploration. By focusing on the 300,000+ digitised image set that is currently available…”

Slide 113

Slide 113 text

but seemed technically achievable in a modern web browser larger than other sets that i had seen

Slide 114

Slide 114 text

little did i know…

Slide 115

Slide 115 text

the scale of the dataset was a big challenge in this project 2 million images…

Slide 116

Slide 116 text

No content

Slide 117

Slide 117 text

emphasis in visual materials 1,199,477 things., photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 118

Slide 118 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 119

Slide 119 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 120

Slide 120 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 121

Slide 121 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 122

Slide 122 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 123

Slide 123 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 124

Slide 124 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 125

Slide 125 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 126

Slide 126 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 127

Slide 127 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 128

Slide 128 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 129

Slide 129 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 130

Slide 130 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 131

Slide 131 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 132

Slide 132 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 133

Slide 133 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 134

Slide 134 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 135

Slide 135 text

1,199,477 manuscripts, photographs, negatives, published maps, drawings, objects, pictures, ephemera, journals, prints, medals, unpublished maps, paintings, coins, architectural drawings, newspapers, posters, and stamps.

Slide 136

Slide 136 text

“unsorted”, year, colour, similarity

Slide 137

Slide 137 text

“unsorted”, year, colour, similarity

Slide 138

Slide 138 text

“unsorted”, year, colour, similarity

Slide 139

Slide 139 text

four different ways of looking at the forest “unsorted”, year, colour, similarity

Slide 140

Slide 140 text

No content

Slide 141

Slide 141 text

No content

Slide 142

Slide 142 text

No content

Slide 143

Slide 143 text

page 1 5 pages

Slide 144

Slide 144 text

8 pages page 1

Slide 145

Slide 145 text

27 pages page 1

Slide 146

Slide 146 text

688 pages page 1

Slide 147

Slide 147 text

27,749 pages

Slide 148

Slide 148 text

No content

Slide 149

Slide 149 text

demo

Slide 150

Slide 150 text

No content

Slide 151

Slide 151 text

No content

Slide 152

Slide 152 text

No content

Slide 153

Slide 153 text

No content

Slide 154

Slide 154 text

No content

Slide 155

Slide 155 text

No content

Slide 156

Slide 156 text

No content

Slide 157

Slide 157 text

No content

Slide 158

Slide 158 text

No content

Slide 159

Slide 159 text

No content

Slide 160

Slide 160 text

No content

Slide 161

Slide 161 text

No content

Slide 162

Slide 162 text

No content

Slide 163

Slide 163 text

No content

Slide 164

Slide 164 text

No content

Slide 165

Slide 165 text

No content

Slide 166

Slide 166 text

No content

Slide 167

Slide 167 text

No content

Slide 168

Slide 168 text

No content

Slide 169

Slide 169 text

No content

Slide 170

Slide 170 text

No content

Slide 171

Slide 171 text

serendipitous exploration

Slide 172

Slide 172 text

No content

Slide 173

Slide 173 text

No content

Slide 174

Slide 174 text

No content

Slide 175

Slide 175 text

unbounded by physical space

Slide 176

Slide 176 text

try this at home… memory limitations apply pushes the limits of web browsers

Slide 177

Slide 177 text

more sorting criteria, custom groupings, filtering from library metadata future work?

Slide 178

Slide 178 text

the process

Slide 179

Slide 179 text

No content

Slide 180

Slide 180 text

The Process of Design Squiggle by Damien Newman, thedesignsquiggle.com

Slide 181

Slide 181 text

ended up using 1.2 million… still a lot for any web browser, let alone a smartphone 2 million images! ⏳

Slide 182

Slide 182 text

No content

Slide 183

Slide 183 text

No content

Slide 184

Slide 184 text

No content

Slide 185

Slide 185 text

No content

Slide 186

Slide 186 text

No content

Slide 187

Slide 187 text

No content

Slide 188

Slide 188 text

No content

Slide 189

Slide 189 text

single pixels can convey a lot of information… but i wanted everything these are all “just” pixels

Slide 190

Slide 190 text

- Frustrated Fellow “but how am i going to show two million images‽”

Slide 191

Slide 191 text

No content

Slide 192

Slide 192 text

No content

Slide 193

Slide 193 text

No content

Slide 194

Slide 194 text

No content

Slide 195

Slide 195 text

No content

Slide 196

Slide 196 text

technical details

Slide 197

Slide 197 text

…with lots of data cleanup and cloud computing time Python + Node.js for image similarity and color palettes

Slide 198

Slide 198 text

cloud-hosted static assets using Library’s API for image details Three.js + Vue.js for Aereo interface

Slide 199

Slide 199 text

github.com/slnsw/dxlab-fellowship-2019

Slide 200

Slide 200 text

dxlab.sl.nsw.gov.au/blog/building-aereo

Slide 201

Slide 201 text

Special thanks to •DX Lab team: Paula Bray, Kaho Cheung, and Luke Dearnley •Web team: Jenna Bain and Robertus Johansyah •Mitchell Librarian Richard Neville •State Librarian Dr John Vallance •Staff from the State Library of New South Wales •The State Library of NSW Foundation

Slide 202

Slide 202 text

Acknowledgments •Cyril Diagne •Douglas Duhaime - Yale DH Lab •Gene Kogan •Mario Klingemann •Ricardo Cabello - Three.js

Slide 203

Slide 203 text

Mauricio Giraldo Arteaga @mgiraldo