매체로서의 트위터, 트위터 텍스트 담론 연구는 지금까지 많이 이루어져왔지만, 해당 매체에서 공유되고 있는 사진 유형에 대해서는 크게 다루지 않았습니다. 본 연구는 미국과 영국에서 작성한 트윗을 임의로 수집, 이와 함께 공유된 사진 800 장을 분석해 평소 사람들이 공유하는 meformer 와 informer 유형의 사진을 구체적으로 살펴보았습니다.

1. Chatting Through Pictures?
A Classification of Images
Tweeted in One Week in
the UK and USA
- Journal of the Association
for Information Science and
Technology, 2016
/ 송지은
+ 2016 Fall

2. Chatting Through Pictures?
A Classification of Images Tweeted
in One Week in the UK and USA
Mike Thelwall, Statistical Cybermetrics Research Group, School of Mathematics and Computer Science, University of Wolverhampton
Olga Goriunova, Centre for Interdisciplinary Methodologies, University of Warwick
Farida Vis, Information School, University of Sheffield
Simon Faulkner, Manchester School of Art, Manchester Metropolitan University
Anne Burns, Information School, University of Sheffield
Jim Aulich, Manchester School of Art, Manchester Metropolitan University
Amalia Mas-Bleda, Institute of Public Goods and Policies, Spanish National Research Council (CSIC)
Emma Stuart, Statistical Cybermetrics Research Group, School of Mathematics and Computer Science, University of Wolverhampton
Franceso D’Orazio, Pulsar, 7 Midford Place

3. Why This Paper
3
/ 사진 연구
/ 저널 알아보기

4. Journal of the Association for Information Science and Technology
4
Impact Factor: 1.864
ISI Journal Citation Reports © Ranking: 2015: 20/86
(Information Science & Library Science); 32/144
(Computer Science Information Systems)

5. INTRODUCTION
5
/ Sharing images through social media is common in richer nations.
These statistics, however,
do not reveal anything about what types of images are shared and why they are shared.
In 2013,
17%of adults on the Internet used Twitter
71% used Facebook,
both of which can be used to share pictures,
16% used the image-sharing site Instagram
(Duggan & Smith, 2013)
In the UK in 2013
70%of adults involved in online activities
reported sharing images
(Dutton, Blank, & Groselj, 2013)
In 2012,
45%americans
posting own photographs online
35%shared photographs
created by others
(52% of those aged 18­29) (Rainie, Brenner, & Purcell, 2012)

6. INTRODUCTION
6
/ Press coverage of Twitter sometimes focuses on celebrity users or on public tweets in reaction to
major events and news stories.
This context will also inform debates
about how image-sharing practices feed into commercial and sociopolitical uses of images on Twitter.
Criado Perez & UK Bank Note Daily Life Photos

7. SOCIAL USES OF IMAGES
7
/ Most Twitter users tweet in a personal capacity, although journalists and media bloggers are also
relatively common, and corporate accounts form a small minority (De Choudhury, Diakopoulos, & Naaman, 2012).
There is no specific evidence to justify this claim, however, and, since most URLs shared in Twitter
originate from a small number of elite media bloggers (Wu, Hofman, Mason, & Watts, 2011), it is possible that most
images shared are also taken or created by an elite minority.
/ Information about how people tweet can help to understand the context of the pictures that they
tweet.
A content analysis of up to 10 tweets each from 350 randomly selected active personal Twitter users
(excluding corporate and self- promoting users) found that the most common types of tweets were the
(overlapping) categories: me now (40%), statements/random thoughts (23%), opinions/complaints
(23%), and information sharing (22%) (Naaman, Boase, & Lai, 2010).

8. SOCIAL USES OF IMAGES 1
8
Personal Image and Photo Sharing (Meformers)
/ The widespread uptake of photography (Beegan, 2008; Benjamin, 1936; Braden, 1983; Marien, 2014) has allowed it to be
embedded in people’s lives through everyday family photographs (Rose, 2010) and for events, such as
weddings and holidays, where the visual is important for long-term memories (Berger, 2013; Cobley & Haeffner,
2009; Robinson & Picard, 2009; Urry & Larsen, 2011).
/ The advent of digital technology has triggered new uses for photography (Gomez Cruz & Meyer, 2012; Miller &
Edwards, 2007; Van House, Davis, Ames, Finn, & Viswanathan, 2005).
/ Photographic images look like pieces of the world as much as statements about it (Sontag, 1978), but
chat can nevertheless occur alongside the images and incorporates them into its primarily verbal
discourse (Hu, Manikonda, & Kambhampati, 2014).

9. SOCIAL USES OF IMAGES 2
9
Information Sharing (Informers): News, Celebrities, and Memes
News Politicians Celebrities
Sports Internet Memes Marketing

10. SOCIAL USES OF IMAGES 3
10
Analyses of Social Media Images
/ Content Analysis of images on Tumblr fan community with texts and hashtags
(Batchen, 2008; Chalfen, 1987; Gomez Cruz & Ardévol, 2013; Holland & Spence, 1991; Larson & Sandbye, 2014; Sarvas & Frohlich, 2011).
Stronger sentiments than image-free posts
Positive emotions (57%) vs. negative emotions (68%)
/ focusing more generally on the content of social media images
(Hu et al., 2014).
A sample of 50 regular active Instagram users and manually coded their most recently posted 20 pictures.
Final set of 8 groups: Pet (3%); Fashion (4%); Food (10%); Gadget (11%); Captioned photo (11%); Activities
(15%); Friends (22%); Selfies (24%)
/ Content Analysis of Photos Shared on Instant Text Messaging between US vs. Taiwanese Users
(Chen, Bentley, Holz, & Xu, 2015)
Images were shared as part of ongoing communications rather than as one-off entities in their own right.
Screenshots were also singled out in this context as a device that allowed complex messages to be
communicated accurately

11. RESEARCH QUESTIONS
11
1. Which types of images are shared on Twitter?
2. What are images shared on Twitter of?
3. Why are these images shared Twitter?
4. When are images shared on Twitter?
and nevertheless…

12. METHODS
12
1,876,484 Tweets, 364, 802 with URLs,
and 196,600 pictures were downloaded
1,484,474 Tweets, 292, 172 with URLs,
and 133.096 pictures were downloaded
/ sampled using the free software Webometric Analyst
/ 7 days from November 29, 2014 at 12:43 GMT
/ blank query was used in conjunction with a geographic restriction
(catching 1% of neighboring countries: Ireland & Mexico)

13. METHODS
13
/ Content Anaylsis:
although not capable of dealing with the nuances of individual images,
it is able to characterize properties of a large set of images in a systematic way,
avoiding at least some of the potential biases of more detailed investigations into small sets of images.
/ Classification Scheme: four main facets
1) the overall type or format of the image (the technical angle)
2) its subject or content ( instagram categories, including selfies)
3) the apparent reason for sharing it (mimetic and celebrities)
4) when it was shared
Image Type of Format Facet Image Content Facet Image Purpose Facet Image Time Relationship Facet
✦ Just photograph. Can have solid black bars
at the top or side; can have a tiny logo or
credit line
✦ M a i n l y p h o t o g r a p h o r c o l l a g e o f
photographs
✦ Image of message(s) only or messages as
the main purpose of
✦ a screen grab
✦ Comic or cartoon
✦ Other
• [extra] Image contains all or part of a
screen grab
• [extra] Image of text or image contains
text that is important to the picture (not
photographer, credit, URL, or Twitter handle)
• [extra] Professional image or content—
apparently taken by a professional
photographer (not amateur pictures taken
for commercial reasons) or made by a
professional graphic designer
✦ Person or parts of person
✦ Small group of people, probably 2­10 people who
are easily individually identifiable
✦ Large group of people, probably 11+ who are not
easily individually identifiable or it is clear that the
group is photo- graphed rather than the
individual members (e.g., from the back, from a
distance)
✦ Animal(s)
✦ Food or drink (including packaged in
supermarket) without a human being
prominently in the picture
✦ Place, such as a stadium, field, house, street, or
town but not a room inside a house
✦ Other things, including messages, song playing
on phone
• [extra] Selfie—photo taken by one of the
people in it even if others or other things are also
prominently in it, not if just a part of the person’s
body
• [extra] TV—is mainly a picture of a TV or a TV
screen or computer used as a TV •
[extra] Pornographic
✦ Advertising a product or service for sale,
including restau- rants or cafes
✦ Joke image. The joke must be inside the
picture, not the tweet text
✦ Event presence—recording presence at a
public event, perfor- mance, sport, meeting,
or similar
✦ Other purpose or purpose not clear
• [extra] Relating in some way to a famous
person other than a cartoon character, or
the artist of any song playing
• [extra] Meme picture or relating clearly to
a memetic use of an image or memetic
image type •
[extra] Christmas-related
✦ • R e a l - t i m e
photograph, screen grab, or artwork that
appears to have been just taken or created
✦ • Current or
topical—not real-time but relating to
something recent or ongoing—such as
advertising for a product, sale, or show
(unless on a poster that has just been
photographed)
✦ • T i m e l e s s ,
including cartoons, memes, and emotional
text messages and sayings
✦ • Old—the picture
is clearly old or posted for historical reasons
(e.g., last year, when we were young)
✦ • Other

14. METHODS
/ Coded manually by two further individuals (PhD in information science)
/ Codes of the first author for 100 UK and 100 US pictures were given to serve as a guide to the
classification scheme.
/ Cohen’s kappa intercoder agreement rates (Cohen, 1960) were then calculated between all three coders
for 597 images (excluding 200 used for training and three images that were difficult to access).

15. RESULTS
15
/ Most of the images were photographs:60%
/displayed text (chat dialogues & sayings): 9%
/ grabs of phones : 15%
/ vast majority of the images did not seem to be
professional
Type of Format
/ People :
individual person: 25%
being of small groups, often just consisting of two
people : 17%
/ Other Category: almost half of the pictures
pictures of text: 20%
images of text or contained text (an overlaid caption):
30%
Contents

16. RESULTS
16
Purpose
/Few tweets revealed a clear reason for
posting an image.
/Some pictures were advertisements or
advertised a product or service.
/ O t h e r c a t e g o r y : s e v e r a l p i c t u r e s
photographed Christmas decorations.
/ Famous people: 15%
/ Few pictures were either a joke or a meme.
Time Relationship
/ Just created : about 70%
which may not always have been true.

17. ANALYSIS OF EMERGENT THEMES
17
✦ 전체 10%가 meformer 전략을 쓰는 것으로 나
타남
✦ 56%이상 혼자 찍고, 42%가 다른 친구들과 찍
음
✦ 한 명 이상과 찍은 사진에서 최소 한 명이 웃고
있었으며 (91%) 혼자 찍은 사진 중 44%가 웃
은 것으로 나타남
✦ 셀카는 대부분 사람이 주가 되었지만, 애완동물
(4%), 장소 (5%), 음식 (3%), 또는 다른 무언가
와 함께 찍기도 함 (4%)
✦ 대부분 손을 멀리 뻗거나, 거울에 비쳐 찍었고 한
명이 셀카봉으로 찍은 것으로 나타남
✦ 그 외 제스쳐: 셀카 8장은 입술을 내밀고 찍었으
며 (여성적인 사진으로 분류), 손으로 브이를 하
고 찍은 셀카가 있었음
✦ 셀카 찍은 사용자는 대부분 여성 (54%)
Selfies Screenshots Texts
✦ 이미지 중 1/3는 이미지에서 중요한 메시지를
담은 텍스트. 절반이 감성적인 메시지였으며,
✦ 이 같은 메시지는 일반 트윗보다 두드러지며,
텍스트의 형태보다 이미지로 보내는 것이 메시
지 전달에 있어 더 효과적임
✦ 이미지 형태는 기본 트윗보다 더 많은 활자를
담을 수 있을 뿐만 아니라 폰트나 스페이싱을
조율할 수 있음
✦ 유사한 목적의 이미지 중에서는 유명한 사람의
사진이 함께 있는 텍스트가 있었음.
✦ 텍스트가 주가 되는 이미지 중에서는
informer 전략을 구사하는 정보의 텍스트가
있었음.
✦ 트위터에서 주로 나타나기도 하는 스크린
샷은 다른 분야 (ex 게임: 본인의 레벨을
달성했음을 기록하기 위해)에서는 많이
다루었지만, 일반적인 소셜 미디어 사진
공유 연구에서 많이 다뤄지지 않음
✦ 스마트폰 스크린샷은 사용자가 무엇을 하
고, 어떤 음악을 들으며 활동하는지를 보
여주므로 meformer 전략으로 분류됨
✦ 이미지 중에서는 informer 전략을 구사
하는 정보의 텍스트가 있었음.
/ The photographs shared primarily depict current content, although perceptions of what is current are
affected by the system used (Weltevrede, Helmond, & Gerlitz, 2014).

18. ANALYSIS OF EMERGENT THEMES
18
이미지 ‘콘텐츠’를 코딩하는 것이 어려움:
1)사진을 보는 사람에 따라 무엇을 강조해서 인식하는지가 다르기 때문
2) 이미지를 찍은 사진(TV screen등) 류의 이미지 분류가 어렵기 때문
사진 공유 목적에 대한 맥락 정보를 제공하지 않음.
대부분의 이미지와 공유가 동시에 발생하는 만큼, 사진을 공유하는 사람이 무엇을 하는지, 보고 있는지, 또는 경험 하는
지 (meformer 전략에 해당하는) 를 공유하는 성향이 있음. 사진 공유의 주요 목적은 따라서 개인의 경험을 면대면 또는
전화에서 확장하여, 상대적으로 간단한 개인의 경험을 친구 또는 주변인들과 교환하는 데 있음.
이미지의 ‘시간 관계’를 파악하는 것도 어려움:
1) 일단은 실시간으로 저장한 사진을 공유할 것이라는 기본 전제를 두었으나 항상 이 같은 추측이 옳을 수 없음
2) 이미지가 ‘몇 초 안에’ 찍혔는지, ‘같은 날 다른 시점’에 찍혔는지 파악하기도 어려움
/ 연구의 어려운 부분

19. LIMITATIONS
19
/ ‘트위터’의 사진 연구에 한해
/ Twitter is not random and is preprocessed for sampling in an unknown way by Twitter
/ the use of Twitter is likely to change over time due to age shifts in its user base, competition from other
services, and new smartphones technologies
/ typical uses are likely to vary substantially over time according to the season

20. CONCUSIONS
20
/ Images shared on Twitter seem to be predominantly photographs used to update friends and
acquaintances about what the user is currently doing
as a visual extension of chatting face-to-face or on social media as a meformer strategy.
/ Images are also used for wider information sharing, although this seems to be a minority activity.
more common to share information related to advertising (7.8%), celebrities (13.9%), or jokes (4.3%)
/ The results suggest several avenues for future research on social media images.
1) it is important to follow up on the apparent use of images for chat or gossip.
2) important to connect and situate findings of image sharing on Twitter to those on other platforms,
and identify similarities and differences in content, topical interests, and users types.

21. 21
End of Document