-
Notifications
You must be signed in to change notification settings - Fork 6
/
chapter3.html
3600 lines (2806 loc) · 321 KB
/
chapter3.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta charset="utf-8" />
<title>Social Web in the Wild</title>
<meta content="width=device-width, initial-scale=1" name="viewport" />
<link href="thesis.css" media="all" rel="stylesheet" title="Thesis do" />
<link href="https://dokie.li/media/css/basic.css" media="all" rel="stylesheet alternate" title="Basic" />
<link href="https://dokie.li/media/css/dokieli.css" media="all" rel="stylesheet" />
<script src="https://dokie.li/scripts/dokieli.js"></script>
<script src="thesis.js"></script>
<link href="https://rhiaro.co.uk/incoming/" rel="http://www.w3.org/ns/ldp#inbox" />
<link href="https://linkedresearch.org/annotation/rhiaro/thesis/" rel="http://www.w3.org/ns/oa#annotationService" />
<style>
h1:before {
content:"Chapter 3\A" !important;
white-space:pre;
font-size:0.75em;
}
/*TODO: Move this to external stylesheet*/
#studies-overview tbody th:first-child { width:200px; }
</style>
</head>
<body about="" prefix="rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns# rdfs: http://www.w3.org/2000/01/rdf-schema# owl: http://www.w3.org/2002/07/owl# xsd: http://www.w3.org/2001/XMLSchema# dcterms: http://purl.org/dc/terms/ dctypes: http://purl.org/dc/dcmitype/ foaf: http://xmlns.com/foaf/0.1/ pimspace: http://www.w3.org/ns/pim/space# cc: https://creativecommons.org/ns# skos: http://www.w3.org/2004/02/skos/core# prov: http://www.w3.org/ns/prov# mem: http://mementoweb.org/ns# qb: http://purl.org/linked-data/cube# schema: http://schema.org/ void: http://rdfs.org/ns/void# rsa: http://www.w3.org/ns/auth/rsa# cert: http://www.w3.org/ns/auth/cert# wgs: http://www.w3.org/2003/01/geo/wgs84_pos# bibo: http://purl.org/ontology/bibo/ sioc: http://rdfs.org/sioc/ns# doap: http://usefulinc.com/ns/doap# dbr: http://dbpedia.org/resource/ dbp: http://dbpedia.org/property/ sio: http://semanticscience.org/resource/ opmw: http://www.opmw.org/ontology/ deo: http://purl.org/spar/deo/ doco: http://purl.org/spar/doco/ cito: http://purl.org/spar/cito/ fabio: http://purl.org/spar/fabio/ oa: http://www.w3.org/ns/oa# as: https://www.w3.org/ns/activitystreams# ldp: http://www.w3.org/ns/ldp# solid: http://www.w3.org/ns/solid/terms# acl: http://www.w3.org/ns/auth/acl# dio: https://w3id.org/dio# rel: https://www.w3.org/ns/iana/link-relations/relation#" typeof="schema:CreativeWork sioc:Post prov:Entity">
<main>
<article about="" typeof="schema:Article">
<h1 property="schema:name doap:name">Social Web in the Wild</h1>
<div id="content">
<section id="introduction" resource="#introduction">
<h2>Introduction</h2>
<div>
<p>The <a href="/thesis/chapter2">previous chapter</a> discusses an individual's relation with their online representation; how users understand profiles; how the affordances of a profile impact the culture of an online community, including how users interact and relate to each other, and how users understand themselves as part of the community. This chapter contains original studies; two which examine online profiles from an outside perspective, by looking at what systems offer and how individuals appear to be making use of this. Three of the studies go behind the scenes to actually ask profile owners about their participation in the social Web ecosystem.</p>
<p>Each study resulted in a small taxonomy useful for categorising the participants' experiences in each particular scenario. A core contribution of this thesis is to coalesce the results of these new studies, along with knowledge from existing literature, into an overall framework consisting of five concepts. This framework - the 5 Cs of Digital Personhood - constitutes the key components for describing online self-expression experiences. The framework is summarised here for reference, and I discuss its derivation in more detail in the conclusion of this chapter.</p>
<figure id="framework-relations">
<img src="figs/20170924_onlinidframework.png" />
<figcaption>Control, Customisability, Connectivity, Context, Cascade; and their relations to each other.</figcaption>
</figure>
<p>Each component encapsulates a variety of different parts or aspects which are revealed through the studies in this chapter, as well as prior research:</p>
<ul id="framework-terms">
<li><strong>Control:</strong> over persistence or ephmerality of identities, attachment or not to real names, traceability between different identities (eg. Can I delete my profile?).</li>
<li><strong>Customisability:</strong> of the data that is included in an online representation, the extent to which this is available to others, and how it is presented (eg. Can I change the name that appears on my profile?).</li>
<li><strong>Connectivity:</strong> to others and an audience, known or imagined, and how impressions by this audience can be managed (eg. Do I know how this profile appears to my mother?).</li>
<li><strong>Context:</strong> the social/cultural expectations of a platform or community; personal motivations and use cases; technical constraints of systems; offline cultural norms or biases which affect or constrain online behaviours (eg. Are the people who control this platform obliged to adhere to the same laws as I am?).</li>
<li><strong>Cascade:</strong> of personal information throughout a network, perhaps unknown; 'profiles' generated by algorithms, data passed around by third parties or collected through surveillance; expression 'given off' over which individuals have little knowledge or control (eg. Is my data being used to recommend products to me?).</li>
</ul>
<p>Whilst all five components influence each other in complex and shifting ways, I illustrate key relations with the following terms: </p>
<ul>
<li><strong>compels</strong>: the existence of aspects of one necessitates the involvement of aspects of the other.</li>
<li><strong>diminishes</strong>: aspects of one reduce the effect of aspects of the other.</li>
<li><strong>enables</strong>: aspects of one increase the effect of aspects of the other.</li>
<li><strong>shapes</strong>: aspects of one feed into aspects of the other; the latter is formed according to or depending on variations in the former.</li>
</ul>
<section id="overview-of-studies" rel="schema:hasPart" resource="#overview-of-studies" style="page-break-before: always">
<h3>Overview of studies</h3>
<div>
<p><a href="#studies-overview">Table 1</a> summarises the methods, inputs and outputs of the five studies in this chapter.</p>
<p>The previous chapter established that there are various different (potentially overlapping) perspectives that need to be taken into account when discussing online self-presentation:</p>
<ul>
<li><b>A</b>ctive users of a system, who maintain a profile.</li>
<li><b>P</b>assive users of a system, who may not have a profile of their own.</li>
<li><b>S</b>ystem designers and developers, who must model and display data about their users.</li>
<li><b>T</b>hird-party developers who build additional services using data from another system.</li>
<li><b>O</b>utside bodies which seek to influence or direct how systems are used for legal, ethical or economic reasons.</li>
</ul>
<p>The five empirical studies in this chapter touch on each of these perspectives to some degree.</p>
<p>The <a href="#what-is-a-profile">first study</a> sets a baseline for describing and categorising online profiles by asking the question "what is a profile?" and takes an objective look at 18 online systems which employ user profiles in a social capacity to classify their features. Subsequent studies focus on the people behind the profiles, or behind the systems themselves.</p>
<p>As hinted at in the previous chapter, individuals are rapidly and often intuitively developing coping mechanisms and practices to improve their handling of online self-presentation and impression management despite the constraints of the tools they use. The studies build on this background, first by observing system users from the outside (in the case of <a href="#constructing-online-identity">creative content producers on YouTube</a>), and then by asking them questions and exploring their feelings and experiences with online profiles, with regards to: <a href="#the-many-dimensions-of-lying-online">deception and lying on social media</a>; <a href="#computationally-mediated-pro-social-deception">imagining social systems as tools for mediating reality</a>; and <a href="#social-media-makers">designing and building one's own customised social systems</a>.</p>
<table id="studies-overview">
<caption>Overview of studies</caption>
<thead>
<tr>
<th>Study</th>
<th>Type</th>
<th>Participants</th>
<th>Publication</th>
<th>Perspectives</th>
<th>Resulting terminology/themes</th>
</tr>
</thead>
<tbody>
<tr>
<td><a href="#what-is-a-profile">What is a profile?</a></td>
<td>Descriptive, observational</td>
<td>18</td>
<td>n/a</td>
<td>A S T</td>
<td>flexibility, access control, prominence, representation, portability</td>
</tr>
<tr>
<td><a href="#constructing-online-identity">Constructing online identity</a></td>
<td>Empirical, observational</td>
<td>10</td>
<td>WWW14</td>
<td>A</td>
<td>roles, attribution, accountability, traceability</td>
</tr>
<tr>
<td><a href="#the-many-dimensions-of-lying-online">The many dimensions of lying online</a></td>
<td>Survey</td>
<td>500</td>
<td>WebSci15</td>
<td>A P S</td>
<td>system, authenticity, audience, safety, play, convenience</td>
</tr>
<tr>
<td><a href="#computationally-mediated-pro-social-deception">Computationally mediated pro-social deception</a></td>
<td>Interviews, design fictions</td>
<td>15</td>
<td>CHI16</td>
<td>A P O</td>
<td>effort & complexity, strategies/channels, privacy & control, authenticity & personas, access & audience, social signalling & empowerment, ethics & morality</td>
</tr>
<tr>
<td><a href="#social-media-makers">#ownYourData</a></td>
<td>Interviews</td>
<td>15</td>
<td>n/a</td>
<td>A S T O</td>
<td>self-expression, persistence/ephemerality, networks & audience, authority, consent</td>
</tr>
</tbody>
<tfoot>
<tr>
<td colspan="1"><strong>Perspectives:</strong></td>
<td colspan="5">A — Active users; P — Passive users; S — System developers; T — Third party developers; O — Outside bodies</td>
</tr>
<tr><th rowspan="4">Publications:</th>
<td colspan="1">WWW14: </td><td colspan="4"><strong>Guy A.</strong> & Klein E. (2014) <em>Constructed Identity and Social Machines: A Case Study in Creative Media Production</em>. Proceedings of the 23rd International Conference Companion on World Wide Web - WWW'14 Companion.</tr>
<tr><td colspan="1">WebSci15: </td><td colspan="4">Van Kleek, M., Murray-Rust D., <strong>Guy A.</strong>, Smith D., O'Hara K., & Shadbolt N. (2015). <em>Self Curation, Social Partitioning, Escaping from Prejudice and Harassment: The Many Dimensions of Lying Online.</em> Proceedings of the ACM Web Science Conference. 10:1-10:9.</td></tr>
<tr><td colspan="1">CHI16: </td><td colspan="4">Van Kleek, M., Murray-Rust D., <strong>Guy A.</strong>, O'Hara K., & Shadbolt N. (2016). <em>Computationally Mediated Pro-Social Deception.</em> Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems. 552–563.</td></tr>
</tfoot>
</table>
</div>
</section>
</div>
</section>
<section id="what-is-a-profile" rel="schema:hasPart">
<h2>What is a profile?</h2>
<div>
<p>This is a descriptive study of 18 systems which employ profiles in a social capacity. This study results in five features and each system is scored according to the degree each feature is present. Using these features we can cluster similar systems together, or differentiate them, and future studies can use these features to create baseline descriptions or characterisations of systems for comparison. The features are: <em>flexibility</em>, <em>access control</em>, <em>prominence</em>, <em>portability</em>, <em>representation</em>.</p>
<section id="profile-intro" rel="schema:hasPart">
<h3>Introduction</h3>
<div>
<p>In order to build on our understanding of the role an online profile plays in self-presentation, identity and interaction we need a more nuanced understanding of what a ‘profile’ <em>is</em> in a general sense. What is the <em>meaning</em> of <em>profile</em>? I carried out an empirical analysis of digital representations of users of 18 different online systems. From this analysis I derive a set of constructs to capture features of profiles in online systems. I propose this for assessing the benefits and drawbacks of how profiles are implemented in existing systems in such a way that takes into account the scenarios in which they are used, as well as groundwork for deriving requirements for profiles when designing new systems which need digital representations of their users. Once we have a characterisation of a particular <em>type</em> of profile a system enables, we can use these as control features when comparing systems side by side. Interesting future study would be to determine how the features of a profile influence actions of users or community formation, and vice versa.</p>
<!-- <p class="notes">Maybe use stuff in participation and interaction of social machines taxonomy http://eprints.soton.ac.uk/362359/1/SOCIAM%20Classificationv2.pdf Could see this as a further breakdown of one of the U categories (see also their methodology pg 13) “Given the value of taxonomies in advancing our understanding of the conceptual landscape associated with a domain”. Taxonomy is very sort of task/activity centric rather than about individuals.</p>
<p class="notes"><a href="https://dud.inf.tu-dresden.de/literatur/Anon_Terminology_v0.21.pdf">Anonymity, Unobservability, Pseudonymity, and Identity Management – A Proposal for Terminology</a></p> -->
<p>For the purposes of this thesis, I define social systems to be Web-based networked publics which offer individuals consistent and reusable access to an account which they can personalise and use to interact in some form with others in the system.</p>
</div>
</section>
<section id="profile-rqs" rel="schema:hasPart">
<h3>Context and research questions</h3>
<div>
<blockquote><p>Profile generation is an explicit act of writing oneself into being in a digital environment (boyd, 2006 [<a href="appendices#ref-boyd06">boyd06</a>])</p></blockquote>
<p>boyd's definition of profile generation above is based on teenagers' use of Friendster and MySpace in 2006. Today, online social systems use profiles in a variety of different ways, and present them in a variety of configurations. Profile generation is not only explicit, but can occur implicitly, without necessarily even the consent or awareness of the profile subject. As discussed in <a href="chapter2">the previous chapter</a>, studies of online profiles tend to focus on oversimplifications or very specific (unrealistic) use cases, which do not take into account the broader system in which the profile exists. This approach often reduces an individual's representation in the system to a single document or webpage, and neglects the rich array of interactions and activities in which they engage in order to create a presence for themselves. In reality, profiles vary in how they are constructed and the roles they play.</p>
<p>This study serves to introduce a formal classification of profile features, and asks the following questions:</p>
<ul>
<li>What are common features of the ways users are represented in online social systems?</li>
<li>How do these features vary between systems?</li>
<!-- <li class="notes">What are potential variants in how profiles are formed which influence a community (and vice versa)?</li> -->
</ul>
<p>Kaplan & Haenlein (2010) categorise social systems into six groups [<a href="appendices#ref-kaplan10">kaplan10</a>]:</p>
<ul>
<li><em>Blogs</em> are "special types of websites that usually display date-stamped entries in reverse chronological order"</li>
<li><em>Social Networking Sites</em> are "applications that enable users to connect by creating personal information profiles, inviting friends and colleagues to have access to those profiles, and sending ... messages between each other"</li>
<li><em>Collaborative Projects</em> "enable the joint and simultaneous creation of content by many end-users"</li>
<li><em>Content Communities</em> are for "the sharing of media content between users"</li>
<li><em>Virtual Gaming Worlds</em> are "platforms that replicate a three-dimensional environment in which users can appear in the form of personalized avatars and interact with each other ... according to strict rules in the context of a massively multiplayer online role-playing game (MMORPG)"</li>
<li><em>Virtual Social Worlds</em> "allows inhabitants to choose their behavior more freely ... there are no rules restricting the range of possible interactions"</li>
</ul>
<p>The subjects of this study (see <a href="#profiles-subjects-table">Table 2</a>) are a cross section of these, but there are also some which do not fit into this framework. Since Kaplan & Haenlein's categorisation, (at least) two new types of system have emerged:</p>
<ul>
<li><em>Quantified Self</em>: life-logging or self-tracking; automated or manual recording of minutiae of daily life;</li>
<li><em>Transactional</em>: networks that exist for exchange of goods or services.</li>
</ul>
</div>
</section>
<section id="profiles-study-design" rel="schema:hasPart">
<h3>Study Design</h3>
<div>
<p>This is a descriptive study [<a href="appendices#ref-grimes02">#</a>], which aims to gather and present information about the current state of social systems with regard to how their users are represented. I do not try to determine causal effects between features of systems, nor do I hypothesise about how these features impact users. Rather, I provide a characterisation of a set of systems as a foundation for future exploratory research.</p>
<section id="profiles-method" rel="schema:hasPart">
<h4>Method</h4>
<div>
<p>I started with the following areas to investigate:</p>
<ul>
<li>Data contained within a profile.</li>
<li>How profile data may be accessed by others (within and outside of the originating system).</li>
<li>How profile data may be distributed or pushed to others (within and outside of the originating system).</li>
<li>The role of profile data within the broader system.</li>
</ul>
<p>The starting point for a 'profile' was typically a unique identifier for an entity (which could be an individual or group) such as a URL or username. After initial explorations of the profiles in a few systems, these areas were refined into specific questions:</p>
<ol>
<li>What does a profile contain?</li>
<li>How are profiles within a system connected together?</li>
<li>How are profiles updated?</li>
<li>How are people notified when a profile is updated?</li>
<li>How is access to a profile controlled?</li>
<li>How can profiles be exported from or imported into a system?</li>
<li>What constraints are placed on a profile?</li>
<li>How do profiles fit in with a systems apparent data model?</li>
<li><em>What</em> is the profile for?</li>
<li><em>Who</em> is the profile for?</li>
</ol>
<p>I took one system at a time, and answered all of the questions by logging in (where applicable) to my own account and observing the behaviours of the system in response to interactions with my own and other users' profiles (where necessary), and took screenshots. I also read systems' terms of service, "About" pages, introductory descriptions or statements of purpose, and leaned on my own background knowledge of how the systems are used by myself and others.</p>
<p>Having answered all of the questions about each system, I passed through each one again to confirm, and add more detail if necessary, and I noted similarities and differences between systems. From the results, I derived a set of potential features for profiles, and ranked each system according to the presence of features. This allowed some clustering of similar systems into a general categorisation framework. </p>
<!--
<p><span class="notes"> Please tell me there's an actual named method for this?</span></p>
<p class="notes"> Evaluation: apply taxonomy to more. Selection how?</p> -->
</div>
</section>
<section id="profiles-subjects" rel="schema:hasPart">
<h4>Subjects</h4>
<div>
<p>18 social systems were selected for the initial analysis phase. Most are ordinary websites which one uses by registering, then logging in and out. Some include or require self-hosted software.</p>
<p>Popular systems which I have personal experience were chosen, in order to take advantage of latent background knowledge when navigating the systems.</p>
<!-- <div class="notes">TODO: Better justify selection of starting 18 sites:
<ul>
<li>Alexa?</li>
<li>Social Machines classifications?</li>
<li>Some dimensions to try to cover, like purpose, user base, business model, ..</li>
</ul>
</div> -->
<p>The information in <a href="#profiles-subjects-table">Table 2</a> serves to give a feel for the diversity of the social systems being studied.</p>
<table id="profiles-subjects-table">
<caption>Profiles study subjects: social systems analysed to generate the taxonomy</caption>
<thead>
<tr>
<th>System</th>
<th>URL</th>
<th>Type</th>
<!-- <th>Est.</th> -->
<th>Specialisation</th>
<th>Overview</th>
<th>Category<sup>k</sup></th>
</tr>
</thead>
<tbody>
<tr>
<td>AirBnb</td>
<td>airbnb.com</td>
<td>website</td>
<!-- <td></td> -->
<td>travelers</td>
<td>Accommodation renting</td>
<td>T</td>
</tr>
<tr>
<td>CouchSurfing</td>
<td>couchsurfing.com</td>
<td>website</td>
<!-- <td></td> -->
<td>travelers</td>
<td>Accommodation, cultural exchange, new connections</td>
<td>T</td>
</tr>
<tr>
<td>Facebook</td>
<td>facebook.com</td>
<td>website</td>
<!-- <td></td> -->
<td>general</td>
<td>New and existing connections</td>
<td>SNS</td>
</tr>
<tr>
<td>Friendica</td>
<td>friendi.ca</td>
<td>website / software</td>
<!-- <td></td> -->
<td>general</td>
<td>New and existing connections</td>
<td>SNS</td>
</tr>
<tr>
<td>Github</td>
<td>github.com</td>
<td>website</td>
<!-- <td></td> -->
<td>developers</td>
<td>Collaborate on software</td>
<td>CP</td>
</tr>
<tr>
<td>Indieweb wiki</td>
<td>indieweb.org</td>
<td>website</td>
<!-- <td></td> -->
<td>developers</td>
<td>Collaborate on ways to develop social web presence</td>
<td>B, CP</td>
</tr>
<tr>
<td>LinkedIn</td>
<td>linkedin.com</td>
<td>website</td>
<!-- <td></td> -->
<td>professional</td>
<td>New and existing connections</td>
<td>SNS</td>
</tr>
<tr>
<td>OkCupid</td>
<td>okcupid.com</td>
<td>website</td>
<!-- <td></td> -->
<td>relationships</td>
<td>New connections</td>
<td>SNS</td>
</tr>
<tr>
<td>PeoplePerHour</td>
<td>peopleperhour.com</td>
<td>website</td>
<!-- <td></td> -->
<td>professional</td>
<td>Hiring freelancers</td>
<td>T</td>
</tr>
<tr>
<td>Pump.io</td>
<td>pump.io</td>
<td>website / software</td>
<!-- <td></td> -->
<td>general</td>
<td>New and existing connections</td>
<td>SNS</td>
</tr>
<tr>
<td>Quora</td>
<td>quora.com</td>
<td>website</td>
<!-- <td></td> -->
<td>general</td>
<td>Q&A (any topic)</td>
<td>CC</td>
</tr>
<tr>
<td>ResearchGate</td>
<td>researchgate.net</td>
<td>website</td>
<!-- <td></td> -->
<td>academic</td>
<td>Advertise/find research publications</td>
<td>CC</td>
</tr>
<tr>
<td>RunKeeper</td>
<td>runkeeper.com</td>
<td>website</td>
<!-- <td></td> -->
<td>sports</td>
<td>Track sporting activities</td>
<td>QS, CC</td>
</tr>
<tr>
<td>StackOverflow</td>
<td>stackoverflow.com</td>
<td>website</td>
<!-- <td></td> -->
<td>developers</td>
<td>Q&A (tech)</td>
<td>CC</td>
</tr>
<tr>
<td>Tumblr</td>
<td>tumblr.com</td>
<td>website</td>
<!-- <td></td> -->
<td>general</td>
<td>New and existing connections</td>
<td>CC, SNS, B</td>
</tr>
<tr>
<td>Twitter</td>
<td>twitter.com</td>
<td>website</td>
<!-- <td></td> -->
<td>general</td>
<td>New and existing connections</td>
<td>B, SNS, CC</td>
</tr>
<tr>
<td>YouTube</td>
<td>youtube.com</td>
<td>website</td>
<!-- <td></td> -->
<td>general</td>
<td>Consume/create media</td>
<td>CC</td>
</tr>
<tr>
<td>Zooniverse</td>
<td>zooniverse.org</td>
<td>website</td>
<!-- <td></td> -->
<td>science</td>
<td>Citizen science</td>
<td>CP</td>
</tr>
</tbody>
<tfoot>
<tr>
<td colspan="7"><strong>Categories from Kaplan & Haenlein:</strong> B — Blog (including Microblog); SNS — Social Networking Site; CP — Collaborative Project; CC — Content Communities</td>
</tr>
<tr>
<td colspan="7"><strong>Additional:</strong> QS — Quantified Self; T — Transactional</td>
</tr>
</tfoot>
</table>
</div>
</section>
<section id="profiles-limitations">
<h4>Limitations</h4>
<div>
<p>As with everything in this thesis, this study is limited by a Western, English-speaking perspective on the systems in question. The observations were conducted from an IP address in either the UK or the US, and I did not attempt to find out how each system differs based on the language preferences or geographical location of users.</p>
<p>Significantly these systems change over time, often rapidly, in response to changing markets, legislation, and available technologies. Most of the data was collected and screenshots captured in the summer of 2015. Some data points were verified to be largely in line with the original findings, but not deeply verified, during writeup in spring 2017. It is important to note that the results are a dated snapshot which cannot be assumed to hold true indefinitely.</p>
<p>I will emphasise again that the nature of a descriptive study does not give any indication of cause-effect relationships between any of the results. Similarly, I can only describe systems as they appear, and not speculate as to <em>why</em> they appear such.</p>
</div>
</section>
</div>
</section>
<section id="profiles-results" rel="schema:hasPart">
<h3>Results</h3>
<div>
<!-- <p class="notes">https://github.com/rhiaro/profiles/blob/master/study.md</p> -->
<p>Here I summarise the findings of the study.<!-- Detailed tables of results are available in <a href="">Appendix X</a>. --></p>
<p>The most distinct of the systems is the Indieweb wiki, which largely functions as an ordinary wiki except that one identifies oneself with a domain name (logging in with the <a href="https://indieweb.org/indieauth">IndieAuth authentication protocol</a>) and thus the 'profile' is tied to one's personal blog, website, or homepage. As a result, profiles are highly custom and diverse; even though they are not hosted centrally by the wiki software they are the main source of identification between users of the wiki, so they are considered here in the same way as the profiles in other systems. In order to study them without visiting the domains of every single user, I also make use of the contents of the wiki itself, which is focused around documenting and recommending best practices for creating a social Web presence; that is, I assume that practices relevant to profile creation described the wiki are adopted by a majority of users.</p>
<section id="profiles-what" rel="schema:hasPart">
<h4>What does a profile contain?</h4>
<div>
<p>Profiles contain some combination of: <strong>attributes</strong> (key-value pairs of data); <strong>content</strong> (text or media) created by the profile owner; a list of <strong>activities</strong> or interactions the profile owner has carried out in the system; <strong>links to profiles</strong> with which they are connected; <strong>links to content</strong> the profile owner has interacted with (e.g. 'likes'); links to <strong>collections</strong> of content curated by the profile owner; <strong>statistics</strong> about the profile (e.g. 'member since'); automatically generated <strong>rankings</strong> or ratings of the profile owner; reviews, messages or content <strong>left by other</strong> members of the network.</p>
<p>All of the 18 systems use attributes in the profile, and none use <em>only</em> attributes. Attributes may be generic (such as <em>name</em>, <em>bio</em>, <em>location</em>), as well as tailored to the specific system (<em>countries I've visited</em> on CouchSurfing; <em>knows about</em> on Quora; <em>looking for</em> on OkCupid). Some attribute values are offered as a fixed set to choose from, and others permit free-text input. Some systems may require a minimum input of certain attributes, and some leave everything entirely optional.</p>
<p>Facebook has the broadest array of possible attributes, including the possibility to create your own keys, and use ones that others have created. CouchSurfing and OkCupid make extensive use of free text input, prompting users to write short essay-style answers to certain questions. Most systems encourage an avatar or display picture, and several also permit uploading a prominent header image (also known as 'banner' or 'cover photo'). The Indieweb community bases attribute-style profile content around the <a href="http://microformats.org/wiki/h-card">microformats h-card</a> specifications, which provides a fixed set, all optional.</p>
<p>Indieweb profiles tend to be the homepages of blogs (although they may be a more static 'about' page) and are heavy on the content and activities aspects. SNS like Twitter, Facebook, Tumblr, Pump.io and Friendica, also lend prominence to content (typically text-based status updates; often photos) and a feed of activities on the site. YouTube incorporates videos created by the profile owner, and how these are organised is highly customisable. For users who have not uploaded video, YouTube profiles contain mostly attributes and activities, and elevate interactions with other content on the site, such as commenting on videos, adding to playlists, and subscribing to channels.</p>
<p>Activity feeds in general vary in their level of detail. Quora displays if someone edited a question or answer. Pump.io distinguishes between 'major' and 'minor' activities, displaying them in separate feeds. Mixed in with posts by the profile owner, Twitter includes a heavily algorithmically curated subset of activities, such as recent follows or likes. Most sites do not include a complete log of all of the possible interactions however. For example, CouchSurfing enables a rich array of activities, from offering to host a guest, to posting in group forums and arranging events; but none of these are displayed on a user's profile. Similarly, most systems do not display a feed of changes to attributes of the profile, which could also be considered activities.</p>
<p>On the other hand, when users interact with content on a system, for example by liking or favouriting it or adding it to a collection (a playlist on YouTube), reblogging it on Tumblr, voting on it on Quora or StackOverflow; this content becomes part of the profile.</p>
<p>StackOverflow, GitHub, PeoplePerHour, ResearchGate, Quora and RunKeeper are very statistics-oriented. RunKeeper focusses on a feed of offline activities, calculating for example how many calories you lost this week from logged exercise, or how far you ran. GitHub visualises code commits and 'contributions' (helpful interactions with projects) in a coloured grid. ResearchGate and Quora display statistics about how much others have interacted with the profile owner's content. OkCupid also generates statistics based on answers to short, multiple-choice personal questions, and these statistics are dependent on who is viewing the profile, e.g. percentage romantic match, and things like '30% more social'.</p>
<p>Sites which make heavy use of content left by others on a profile are CouchSurfing, AirBnB, and PeoplePerHour. Each of these display reviews of the profile owner by other users, typically in a way that cannot be amended or removed. Facebook allows one to 'write on the wall' of another profile, but users can disable this. However, comments and likes by other users commonly show up alongside activities or created content on a profile as well. LinkedIn prompts users to 'endorse' one another for particular skills, and these endorsements are prominent on profiles. StackOverflow and Quora aggregate ratings left by others on content into overall numbers or rankings to display on profiles.</p>
<p>Many systems give prominence to the connections with other users in the system; LinkedIn displays neither likes nor status updates on the profile, but emphasises contacts and the network around them; Twitter displays followers and following; YouTube, ResearchGate, Pump.io, Friendica, and Quora display subscriptions and subscribers.</p>
<!--<pre class="notes">**Indieweb**: Attributes (fixed as per microformats2 but extensible), authored content.
**Twitter**: Attributes (fixed set), connections (subset), authored content, activities within the network (heavily curated).
**Tumblr**: Content; original posts and reposts. Attributes: username, blog title, blog description. (Separate page for likes which is optionally public). No stats by default.
**Pump.io**: Attributes (name, hometown, bio, avatar) and content (major in main feed) and activities (minor down the side) (ActivityStreams)
**YouTube**: channel art, display picture, content (videos), liked videos, created playlists, curated connections, featured content. Links to: attributes (description, links, country); stats (subs, views, joined); discussion (comments on your content).
**CouchSurfing**: Attributes (name, age, location, occupation, education, hometown, languages. member since, photo, availability), free text (about me, interests, music movies & books, one amazing thing done, teach learn share, what I can share, countries visited, countries lived in), % complete profile (pushes to connect to fb), verified; links to My Home, Photos, References, Friends.
**Facebook**: Attributes (literally everything you can think of, free text and enum and select from content created by others e.g. pages). Content (status updates, notes, photos, videos, other media). Stats (friends, following, likes, groups). Activities (interactions with other people and content, stuff cross posted from linked apps and sites).
**Quora**: Attributes (name, photo, bio, description, social media links, location, knows about); Activity (questions, answers, votes, comments, edits, posts); stats (# ^, # followers, # following, highlights (badges basically), views on answers)
**Okcupid**: Attributes (photo, age, location, gender, Self-summary, What I'm doing with my life, I'm really good at, Favorite books, movies, shows, music and food; The six things I could never do without; The first thing people usually notice about me; I spend a lot of time thinking about; On a typical Friday night I am; You should message me if; I'm looking for [Men/Women/Everyone], ages between, near me?, must be single?, for [new friends / long-term dating / short-term dating / casual sex]; orientation; ethnicity; status; relationship type; height; body type; diet; smoking; drinking; drugs; religion; sign; education; offspring; pets; languages spoken). Stats (personality traits compared to average; % match with person reading profile). Content (photos, questions and answers).
**StackOverflow**: Attributes (name, role, company, about me, location, links and social media); stats (answers, questions, people reached, member for, profile views, last seen, visits, badges, accept rate, vote counts); content (posts q&q, tags); links to activity, reputation, everything done.
**LinkedIn**: Attributes (name, photo, location, hometown, experience, education, summary, languages, voluntary experience, organisations, honors and awards, test scores, courses, patents, projects, publications, certifications, interests, personal details, contact information, skills, background photo); Content (posts, media) (not on profile: likes, status updates); Stats (groups, following, rank, views); from others (endorsements).
**PeoplePerHour**: Attributes (name, job title, cover image, profile picture, per hour rate, phone, about you, skills (enum), location, remote only or onsite possible, miles willing to travel. Content (intro video; portfolio files). Stats (# reviews, online presence, rating, projects completed, buyers worked with, last project). From others (reviews, endorsements).
**ResearchGate**: Attributes (name, institution, discipline and subject(s), skills, photo, email, topics, degree, position, awards and achievements, about, advisors, scientific societies, ORCHID, journal referree, interests. Content (publications, many types). Stats (# publications, reads, citations, impact points, top co-authors, following, followers).
**AirBnB**: Attributes (name, location, ID type, school, languages, photo, about; private by default: email, phone, gender, date of birth). Stats: member since, # reviews, (un)verified). Others (reviews). Content ('symbol', profile video 30 secs).
**Friendica**: Attributes (photo, relationship status, sexual preference, display name, name, gender, birthday, homepage, birthday, address, location, public keywords, bio (free text), a bunch of other free text fields for interests etc). Content (status, photos, videos, personal notes). Activity (shares, likes, dislikes, stars, events, rsvps). Stats (member since). Others (messages).
**Zooniverse**: Attributes (name, display name, photo, header image); Activities (favourites, contributions); Content (comments, collections).
**runkeeper**: Attributes (name, photo, location, motivation, gender, birthday, weight, body measurements, nutrition, certain health conditions); Activities (recent PRs, summaries, exercise tracked); Stats (miles, # activities, calories, # friends, active since)
**github**: Attributes (name, location, email, website, bio, photo, company, available for hire?); Activity (code contributions, issues, comments); stats (joined, followers, starred, following, organisation membership)</pre>-->
</div>
</section>
<section id="profiles-connections" rel="schema:hasPart">
<h4>How are profiles within a system connected together?</h4>
<div>
<p>Connections between profiles may be uni- or bi-directional; some systems permit both. <strong>Bi-directional</strong> connections need to be mutual; triggered by one user and confirmed by the second. <strong>Uni-directional</strong> connections may or may not need approval from the second user, depending on either the system as a whole or individual user preferences. Some systems contain more than one kind of uni-directional connection, which may be named or displayed differently, and carry different connotations. Systems vary in whether or not they notify other users (than the ones involved in the connection) about new connections.</p>
<p>Systems with uni-directional connections are Twitter, Tumblr, Pump.io, Facebook, Quora, LinkedIn, ResearchGate, Friendca and GitHub ('follow'), YouTube ('subscribe'), OkCupid, PeoplePerHour ('like/bookmark/favourite'). Systems with bi-directional connections are CouchSurfing, Facbook, Friendica, and RunKeeper ('friends'), LinkedIn ('connect'). The intersection of these (systems with both) is Facebook, LinkedIn, and Friendica.</p>
<p>Some Indieweb profiles include a list of others the profile owner follows using XFN markup [<a href="appendices#ref-xfn">xfn</a>], but this is not necessarily widespread. StackOverflow, Zooniverse and AirBnb do not have a means of creating persistent connections between profiles, besides leaving references in the case of AirBnb.</p>
<p>Systems which permit more specific information or <strong>categorisation of connections</strong> are CouchSurfing (specify 'hosted', 'surfed', 'traveled with' or 'never met' as well as the closeness of the relationship), and Facebook (can specify specific relationships, e.g. 'brother'). When a follow request is sent on Friendica, the recipient can accept it as uni-directional (the follower is labelled a 'fan/admirer') or bi-directional, so the recipient also sees the follower's updates. Bi-directional connections on LinkedIn require a reason or more information as 'proof' of a mutual connection, before the request is even sent.</p>
<p>YouTube connects profiles together through subscriptions to channels, however it also explicitly provides input for profile owners to link to other profiles without creating a subscriber relationship. This lets content creators list, for example, other users they admire, or the people they collaborate with. Many YouTubers use this feature to link to other profiles they have on the site. The system gives users free text fields to name this list, as well as each individual link in the list. This particular phenomenon is examined in more detail in the next study, Constructing Online Identity.</p>
<p>OkCupid connections are uni-directional, and only revealed to the recipient if and when a mutual action is made. On Twitter, following another user sometimes (not consistently) appears as an activity in your timeline; notifications are also sometimes sent to your followers to advertise the new connection.</p>
<!--<pre class="notes">**Indieweb**: Under debate, some people use XFN (e.g. `rel=following`)
**Twitter**: One directional; notifications sometimes sent to your followers when you follow someone, especially if others in your network also do; follows (some? all? curated how?) appear in your timeline; no notifications of unfollows; configurable notification when someone follows you.
**Tumblr**: `Follow` a blog. (Not a user).
**Pump.io**: `Follow` a user. Minor activity and notification when someone follows you. Added to 'follows' collection (public). One way.
**YouTube**: `Subscribe` to a channel, one way.
**CouchSurfing**: Friends — specify hosted/surft/traveled/never met, and closeness. Couch request/offer (but this doesn't create a persistant connection). References.
**Facebook**: Friends — mutual, accept/reject request. You can choose to let anyone 'follow' your public posts (or not). You can follow pages. Following is one way.
**Quora**: follow a user (one way)
**Okcupid**: Like/bookmark (same thing, interchangeable language). Only revealed if mutual.
**StackOverflow**: none? RSS feeds for user activity..
**LinkedIn**: Follow (one way) someone if they allow. 2 way 'Connect' can require reason or input contact details as proof.
**PeoplePerHour**: Add user to favourites.
**ResearchGate**: Follow (one way).
**AirBnB**: n/a (only references, nothing persistant)
**Friendica**: Follow request. Reciever gets to accept as one way (fan/admirer; they read you) or two way (friend; you also see updates from them).
**Zooniverse**: n/a
**runkeeper**: 2 way friendship (request-accept); used with access control as 4, and delivery/notifications.
**github**: follow a user (see all activity in feed), but primary method seems to be to follow content/threads.</pre>-->
</div>
</section>
<section id="profiles-updates" rel="schema:hasPart">
<h4>How are profiles updated?</h4>
<div>
<p>Profiles may be updated by profile owners via a system's user interface, programmatically through an API (Application Programming Interface; the means through which data can be read or written by third-party software). <!--I posit that the options presented for modifying a profile can influence how users make use of a system for their self-presentation.--> The latter is relevant because programmatic access suggests that third-party applications (outside of direct control the system itself) can also influence a profile owner's view on the possibilities of the profile.</p>
<p>Most systems provide a <strong>Web form</strong> to add or update <em>attributes</em>, or a similar UI in a native mobile application.
The editing interface and the profile display may be tightly coupled (Twitter, Quora, LinkedIn, ResearchGate) completely divorced, or a combination (Facebook, OkCupid). Indieweb profiles are updated with custom editing interfaces, or simply by editing static HTML; there are currently no specific recommendations for protocols or UIs to edit profile attributes.</p>
<p>For the non-attribute data which makes up a profile, separate, often <strong>specialised interfaces</strong> for both Web and mobile exist, e.g. for posting status updates or media content. For data like statistics and activities, this content is generated by <strong>algorithms or sensors</strong>, with no explicit input from the profile owner. In a few cases it may be hidden by the profile owner, but rarely changed. An exception is RunKeeper, where one can edit an automatically generated GPS trace after the fact, which can correct distance and speed records. On CouchSurfing, AirBnB and PeoplePerHour, one may respond to a review left by someone else, but not remove it.</p>
<p>Only Pump.io, RunKeeper and GitHub provide <strong>APIs</strong> to update all <em>attributes</em> of a profile. Facebook and Zooniverse provide limited update access to certain attributes. Most systems provide write APIs to create, follow and like (or equivalent) non-attribute content.</p>
<!--<pre class="notes">**Indieweb**: not spec'd (likely editing static HTML).
**Twitter**: Web form / client UI.
API:
WAIT THESE ARE ALL GET. No POST for update o.O
* `/user/info`: following, default_post_format, name, likes, blogs [name, url, title, primary (bool), followers, tweet (bool), facebook (bool), type (pub/priv)]
* `/user/dashboard` (which is feed of what they see that they're subscribed to etc)
* `/user/likes`, `/user/following`
**Tumblr**: Web form, mixed in with blog appearance/settings.
**Pump.io**: Web form / client UI
API:
<q>`
To create a new user, POST a user representation (see below) to the list.
The JSON object representing the user has the following properties:
* nickname: The user's nickname. 1-64 characters, including only ASCII capital and lowercase letters and numbers as well as "-", ".", and "_". The nickname is immutable and unique per server; it can't be changed.
* password: The plain-text password. This isn't returned when you GET the user object, but you have to provide it when registering or updating the user.
* profile: a "person" object. This is created automatically when you create a new user; don't try to add it yourself. Don't update this directly; update the person through its object endpoint.
<q>`
**YouTube**: Web form(s) / client UI
**CouchSurfing**: Web form
**Facebook**: Various web forms, some inline.
<q>`
You can update a User by making a POST request to /{user_id}.
id
firstname
lastname
label_cohort
<q>`
Also various endpoints for types of content that might appear on profiles.
**Quora**: Web form, built into profile view.
**Okcupid**: Web forms, some inline, some separate.
**StackOverflow**: Web form
**LinkedIn**: Web form built into profile view.
**PeoplePerHour**: Web form
**ResearchGate**: Web form built into profile view.
**AirBnB**: Web form.
**Friendica**: Web form (API for profile updates not implemented)
**Zooniverse**: Web form.
API: The currently logged in User may edit their record by sending a partial representation of the resource including their changes. A User cannot edit linked resources:
<q>`
PUT https://panoptes-staging.zooniverse.org/api/users/{id}
Accept:application/vnd.api+json; version=1
Content-Type:application/json
{
"users": {
"credited_name": "Dr. Stuart Lynn, DDS"
}
}
<q>`
**runkeeper**: Web form. API:
<q>`
PUT /profile HTTP/1.1
Host: api.runkeeper.com
Authorization: Bearer xxxxxxxxxxxxxxxx
Content-Type: application/vnd.com.runkeeper.Profile+json
{
"athlete_type": "Ultra Marathoner"
}
<q>`
**github**: Web form. `PATCH /user`: name, email, blog, company, location, hireable, bio</pre>-->
</div>
</section>
<section id="profiles-notifications" rel="schema:hasPart">
<h4>How are people notified when a profile is updated?</h4>
<div>
<p>The attention a system draws to profile updates could affect how people engage with their own profiles. When profile attributes are updated by the profile owner, most systems <strong>do not notify</strong> other users of the system at all.</p>
<p>Facebook however pushes updates to friends' timelines along with status updates and content interactions, though the extent to which it does this for each friend depends on their <strong>arbitrary</strong> content distribution algorithm, and from a user perspective is hard to predict. The most reliably seen attribute updates are changes to profile pictures, cover photos, and relationship status. Whenever the profile owner updates an attribute on Facebook, they are asked to make it a 'story', which sustains a reference to the fact the attribute changed. Friendica notifies about changes to profile pictures only.</p>
<p>OkCupid and LinkedIn provide the <strong>option to enable sharing of changes</strong> to profile attributes. In the case of LinkedIn, updates are pushed to contacts' feeds, but may also be displayed to non-immediate contacts in the network as a form of promoting connections. OkCupid may display updates to other users in their activity feeds according to whether the system thinks these people might be interested in your profile. How either of these are decided is opaque to the user.</p>
<!--<pre class="notes">**Indieweb**: not spec'd
**Twitter**: None.
**Tumblr**: None.
**Pump.io**: minor 'update' activity
**YouTube**: None.
**CouchSurfing**: None.
**Facebook**: Some (most) attributes in timeline; particularly profile pics, cover photos, relationship status (except when option not to publish is chosen, per thing). When you update attributes it asks if you want them to be a story. Some things as life events.
**Quora**: None.
**Okcupid**: Yes, in your activity stream (people wh you looked at or in any way interacted with or who okc just decided to show you).
**StackOverflow**: None.
**LinkedIn**: If enabled by subject.
**PeoplePerHour**: None.
**ResearchGate**: None.
**AirBnB**: None.
**Friendica**: Profile pic updates.
**Zooniverse**: None.
**runkeeper**: None.
**github**: None.</pre>-->
</div>
</section>
<section id="profiles-access" rel="schema:hasPart">
<h4>How is access to a profile controlled?</h4>
<div>
<p>Systems may provide all-or-nothing access to profiles, make everything public but all optional, provide access control on the basis of groups or networks, or individual users, and provide granular access to individual aspects of profiles.</p>
<p>Systems which have limited or <strong>no access control</strong>, but make all or most data <strong>optional</strong> to enter include OkCupid, Quora, CouchSurfing, AirBnB, Friendica, Zooniverse, Pump.io and GitHub. OkCupid and CouchSurfing allow profile visibility to be restricted to other logged-in users. CouchSurfing permits users to hide their full name, and GitHub permits users to hide their email address.</p>
<p>Quora permits users to answer or ask questions as <strong>'anonymous'</strong> whilst logged into their account. These questions/answers do not show up on the user's profile. Otherwise, the only other control profile owners have is disabling their online presence. Friendica permits connections to be hidden, as well as certain aspects of content. On AirBnB, profile attributes are optional but <em>hosts</em> can automatically decline users who omit certain attributes.</p>
<p>Systems with more <strong>granular concepts of audience</strong> than public/private include Pump.io, LinkedIn, Facebook, Twitter, and ResearchGate. In Pump.io individual objects can be 'addressed' so that only particular groups (which can be created by the profile owner) or individuals can see them. LinkedIn permits visibility of some individual profile attributes to 'everyone', 'my network' and 'my connections'. The profile can be set to publicly visible, with certain attributes individually excluded. Connections can be private or public, and content and interactions can be designated different levels of visibility from entirely private to entirely public, with 'network' and 'connections' in between. ResearchGate enables hiding certain statistics, certain attributes, and certain content. Uploaded papers can be visible to 'everyone', 'mutual followers' or 'ResearchGate members'.</p>
<p>Twitter allows users to 'protect' their profiles, which means only those <strong>requesting access</strong> can see content and connections; however, all <em>attributes</em> are visible to anyone regardless. Profile owners can block other users, preventing them from seeing everything but their name, display picture and profile banner.</p>
<p>Systems with granular access control across several different aspects of the profile include YouTube, Facebook, RunKeeper and ResearchGate. YouTube provides granular access controls for various attributes, interactions, links to content, some statistics (like number of subscriptions) and content. RunKeeper attributes can be assigned levels of visibility individually ('everyone', 'friends', 'just me').</p>
<p>Facebook has complex granular access controls, including individual attributes, content, interactions, connections and links. Defaults can be set, as well as updated on a per-object basis at the time of posting/creating. Content can be restricted to include or exclude individuals, groups, particular networks. Read and write access controls are distinct; that is, one can create a post that is publicly readable, but comments on that post may be restricted or disabled completely.</p>
<p>Tumblr's use of 'primary' and 'secondary' blogs is interesting; where a blog constitutes a profile, users can essentially have as many profiles as they want attached to one login. Primary blogs (one per login) are always public, but secondary blogs (unlimited) can be <strong>password protected</strong>. There are no automatic <strong>links between</strong> a user's primary blog and secondary ones, including through the API. There is also no way to tell if a particular profile is primary or secondary, or the account to which a secondary blog is attached. Secondary blog owners may also grant <em>write</em> access to other system users, enabling multi-user profiles. Blocking users prevents the blocked user from interacting with or seeing content.</p>
<!--<pre class="notes">**Indieweb**: not spec'd
**Twitter**: All or nothing for content. A blocked user can still see some attributes (avatar, banner, name). Protected accounts show all profile data but not content or followers/ing
**Tumblr**: "a primary blog is public facing", secondary blogs can be password protected. Per-blog basis.
No automatic links between primary and secondary blogs, and no way to tell if a blog is primary or secondary, or who which user owns a secondary blog (even from API without trial and error).
You can block users: they can't follow, send fan mails or asks, see your posts in their dashboard, like, reblog or reply to your posts, see you in search results.
Can grant write access and transfer admin between users for secondary blogs.
**Pump.io**: Individual resources can be addressed to groups or users, but profile attributes are all public (but none required). No anonymous activities. No way to invisibly do minor activities.
**YouTube**: Pretty granular. pub/priv: liked videos, saved playlists, subscriptions; subscribers. exist or not: name, country, keywords. Appear as a recommended channel on/off.
**CouchSurfing**: Most things optional. Options: First or full name; restrict to log in members.
**Facebook**: Can restrict access to individual attributes on profile. Also per-post. Can set defaults to include or exclude people/groups/public/networks. Super complex. Can restrict findability of profile. Can set commeting access on public posts and profile info separately (ie. read and write acl are separate).
**Quora**: Online presence on/off. Nothing for profile attributes. Can choose to ask/answer as Anonymous, but still logged in.
**Okcupid**: Requires okc login on/off. Nothing required.
**StackOverflow**: Most fields optional, not granular visibility.
**LinkedIn**: Some individual attributes visibility: My Connections / My Network / Everyone. Public profile visible to: No-one / Everyone with on/off for (Basics, Picture, Headline, Websites, Summary, Skills, Organisations, Groups). Connections: Connections / Only You. Public posts: Everyone / Connections. Activity feed: Everyone / Network / Connections / Only You.
**PeoplePerHour**: Show hourlies on/off. Index by search engines on/off.
**ResearchGate**: Research data: everyone / mutual followers / researchgate members. If you've read other peoples' on/off. Fixed on/off settings for some things: public profile, photo, questions and answers.
**AirBnB**: Some things are private by default. Social connections to share activity on fb only with other airbnb users, and suck some data through from fb. Hosts can require certain profile attributes are filled in.
**Friendica**: Tags have separate fields for public/private. Friends/contacts visibility on/off. Everything optional. "This is your public profile. It may be visible to anybody using the internet."
**Zooniverse**: All optional.
**runkeeper**: 'Sharing' Everyone/Friends/Just Me, profile things in groups of different levels of granularity (body measurements, weight, diabetes, nutrition, sleep, activities). Anything not in this list is an optional thing to fill in I guess, ie no way to say 'name' is friends only.
**github**: Most optional, can choose to keep email private.</pre>-->
</div>
</section>
<section id="profiles-portability" rel="schema:hasPart">
<h4>How can profiles be exported from or imported into a system?</h4>
<div>
<p>In the Indieweb model of profile ownership, all data is assumed to be on a server controlled, or at least trusted, by the profile owner. As such, they can move it however they please. Similarly, Pump.io and Friendica are open source software platforms which allow people to either opt to use an instance on a server they trust, or install their own instance for complete control. They both use the standard ActivityStreams 1.0 data model [<a href="appendices#ref-as1">as1</a>] (Friendica has extensions); while Friendica provides <strong>import/export</strong> functionality in the UI, Pump.io doesn't, however the database or JSON feed is compatible across instances.</p>
<p>Twitter, Facebook, YouTube, LinkedIn, ResearchGate, and RunKeeper provide a <strong>download link</strong> for an archive of content. In most cases these are a snapshot of current profile attributes, without a history of changes, except for Facebook, which provides a comprehensive activity log. All exports are proprietary schema in JSON, HTML or CSV.</p>
<p>StackOverflow profiles are <strong>reusable</strong> across different StackExchange sites; there is no export, however there are public dumps of Q&A data. GitHub data is available through an <strong>API</strong>.</p>
<p>Tumblr, CouchSurfing, Quora, OkCupid, PeoplePerHour, AirBnB and Zooniverse provide neither an export nor an API to access all profile data.</p>
<!--<pre class="notes">**Indieweb**: Server agnostic, move data as you please, but you need to control your domain.
**Twitter**: You can download an archive of content and snapshot of profile attributes (and date of snapshot) but no changes history.
**Tumblr**: Third party tools only, references from some time ago, don't know if they still work.
**Pump.io**: Open format, AS JSON. No API for export, or anything in Web client. Can theoretically move between servers.
**YouTube**: Google Takeout export as JSON or OPML. Export analytics.
**CouchSurfing**: n/a
**Facebook**: You can download all of your data. https://www.facebook.com/help/405183566203254 "This includes a lot of the same information available to you in your account and activity log, including your Timeline info, posts you have shared, messages, photos and more. Additionally, it includes information that is not available simply by logging into your account, like the ads you have clicked on, data like the IP addresses that are logged when you log into or out of Facebook, and more."
**Quora**: No export, no API
**Okcupid**: No official API, no export.
**StackOverflow**: Profiles reusable across StackExchange sites. Export n/a through UI, but there are public dumps of q&a data.
**LinkedIn**: Can download a full archive of all data, activity, connections, *everything*.
**PeoplePerHour**: n/a
**ResearchGate**: Export profile as csv.
**AirBnB**: Nothing for profile. Export csv of transaction history, and various calendar syncing.
**Friendica**: ActivityStreams. Various levels of compatability with APIs of GNU Social, pump.io, Twitter, diaspora. Profile and connections portable to another server. You can also use your friendica identity via webfinger to identify with a bunch of other things like remotestorage, owncloud.
<q>`
* Friendica uses DFRN as message protocol.
* The messages uses ActivityStreams.
* Webfinger is used for the discovery of users
* Salmon is used as a message exchange protocol for replies and mentions.
* Portable Contacts is used for friends lists.
* pubsubhubbub is used for OStatus.
<q>`
**Zooniverse**: n/a
**runkeeper**: All data exportable, not standard format.
**github**: repo content is inherantly portable via git; other stuff avaiable through API but no standard format.</pre>-->
</div>
</section>
<section id="profiles-constraints" rel="schema:hasPart">
<h4>What constraints are placed on a profile?</h4>
<div>
<p>In this section I examine the terms of service of systems to determine how users are <em>expected</em> to engage. In some cases these are enforced by technical constraints.</p>
<p>Twitter, CouchSurfing, Facebook, OkCupid, LinkedIn, PeoplePerHour, AirBnB and GitHub state that a user <strong>may not have multiple</strong> accounts. Twitter qualifies this with "overlapping use cases".</p>
<p>Tumblr users cannot create two primary blogs with the <strong>same email address</strong>, and can create 10 secondary blogs per day on the same login with no overall limit. Secondary blogs are somewhat constrained in their functionality compared to primary blogs.</p>
<p>Couchsuring, Facebook, Quora, StackOverflow, LinkedIn, PeoplePerHour, AirBnB, GitHub and RunKeeper explicitly <strong>disallow 'fake'</strong> profiles; the profile owner must be a single 'real' person, and <strong>not be impersonating</strong> someone else.</p>
<!--<pre class="notes">**Indieweb**: n/a
**Twitter**: ~"may not create multiple accounts with overlapping use cases"
**Tumblr**: Needs valid email. Create 10 secondary blogs per day. No way to have two primary blogs on the same email address.
Secondary blogs can't 'initiate social features' (follow, like, ask, fan mail etc).
**Pump.io**: Server specific ToS?
**YouTube**: Some ToS rules about linking to external sites.
**CouchSurfing**:
<q>`
5. Do Create Only One Profile: Duplicate, fake, and joke profiles are not allowed. The first profile that you create must be you and is the only one that you may have. Our trust network needs everyone to stand by their reputation.
6. Do Be Yourself: Misrepresenting yourself as someone else is prohibited. This includes representation as an agent, representative, employee, or affiliate of Couchsurfing.
<q>`
**Facebook**:
<q>`
Registration and Account Security
Facebook users provide their real names and information, and we need your help to keep it that way. Here are some commitments you make to us relating to registering and maintaining the security of your account:
* You will not provide any false personal information on Facebook, or create an account for anyone other than yourself without permission.
* You will not create more than one personal account.
* If we disable your account, you will not create another one without our permission.
...
* You will keep your contact information accurate and up-to-date.
* You will not share your password (or in the case of developers, your secret key), let anyone else access your account, or do anything else that might jeopardize the security of your account.
* You will not transfer your account (including any Page or application you administer) to anyone without first getting our written permission.
* If you select a username or similar identifier for your account or Page, we reserve the right to remove or reclaim it if we believe it is appropriate (such as when a trademark owner complains about a username that does not closely **relate to a user's actual name**).
<q>`
**Quora**:
<q>`
... you must provide us accurate information, including your real name, when you create your account on Quora.
agree that you have not and will not contribute any Content that ... (e) creates an impression that you know is incorrect, misleading, or deceptive, including by impersonating others or otherwise misrepresenting your affiliation with a person or entity;
<q>`
**OkCupid**:
<q>`
Unique and bona fide profile
You agree to create only one unique profile. In addition, in order to maintain the integrity of the Website, by joining, you agree that your use of the Website shall be for bona fide relationship-seeking purposes (for example, you may not use the Website solely to compile a report of compatible singles in your area, or to write a school research paper). From time to time, we may create test profiles in order to monitor the operation of our services.
<q>`
**StackOverflow**:
<q>`
Subscriber shall provide Stack Exchange with accurate, complete, and updated registration information, including Subscriber’s e-mail address. Failure to do so shall constitute a breach of this Agreement, which may result in immediate termination of Subscriber's account. Subscriber may not (a) select or use as a profile name a name of another person with the intent to impersonate that person; or (b) use as a profile name a name subject to any rights of a person other than Subscriber without appropriate authorization.
<q>`
**LinkedIn**:
<q>`
To use the Services, you agree that: ... (2) you will only have one LinkedIn account, which must be in your real name
You agree to ... (3) not transfer any part of your account (e.g., connections, groups)
<q>`
**PeoplePerHour**:
<q>`
To use PPH, Users have to register for an account, provide accurate and complete information, and keep their account information updated. Both Buyers and Sellers undergo the same account registration process.
Each account must be a personal account, but Users may trade as a sole trader, company or any other legal entity (whether incorporated or unincorporated).
Users cannot register for more than one account.
The information that Users supply in the public profile of their account must comply with PPH profile policies (http://support.peopleperhour.com/hc/en-us/articles/205218177-Profile-policies)
Your profile is your “shop front” as a Seller so it is recommended that you invest the time to include as much information as you can to highlight your skills and credentials. When creating your profile we have a few policies that Sellers are expected to adhere to. These policies form part of our T&Cs.
Do
make sure that all of the information contained in your profile is a true and accurate representation. Including your name, location and details of your credentials;
include a professional image of yourself as your profile image. You can choose to have your company logo instead however we strongly recommend a personal photo instead. The PPH marketplace is built on trust and personal photos rather than logos have been shown to make a significant difference to success;
include as profile cover a quality image or photo of yourself or a scenery. Avoid excessive amount of text keeping it to the length of a tagline. You can include visual representation of your work or part of your actual portfolio if visual.
Don't
include your full name, or contact details in your profile description, cover image, portfolio items or any other section of your profile. This rule is first and foremost to protect you; we would discourage you from publishing your contact details visibly on any public site.
open more than one user account with PeoplePerHour. Duplicate accounts will be closed;
include any links to your website or any third party website anywhere in your profile including your profile image or portfolio images;
upload any content to your profile that breaches our terms and conditions including all the obvious things like abusive, offensive, defamatory or infringing content.
share your bank account or any other withdrawal method. Such practice is against our Terms and Conditions of Website Use and leads to permanent account suspension.
<q>`
**ResearchGate**: Nothing to do with profiles?!
**AirBnB**:
<q>`
Your Airbnb Account and your Airbnb Account profile page will be created for your use of the Site and Application based upon the personal information you provide to us or that we obtain via an SNS as described above. You may not have more than one (1) active Airbnb Account. You agree to provide accurate, current and complete information during the registration process and to update such information to keep it accurate, current and complete. Airbnb reserves the right to suspend or terminate your Airbnb Account and your access to the Site, Application and Services if you create more than one (1) Airbnb Account, or if any information provided during the registration process or thereafter proves to be inaccurate, fraudulent, not current or incomplete.
you will not: ...
* register for more than one Airbnb Account or register for an Airbnb Account on behalf of an individual other than yourself;
* unless Airbnb explicitly permits otherwise, request or book a stay at any Accommodation if you will not actually be staying at the Accommodation yourself;
* impersonate any person or entity, or falsify or otherwise misrepresent yourself or your affiliation with any person or entity;
<q>`
**Friendica**: ToS per server presumably.
**Zooniverse**: Nothing profile related
**runkeeper**:
<q>`
agree that you will not: ... (v) impersonate another person.
<q>`
**github**: