Why We Tag: Motivations for Annotation

in Mobile and Online Media

Morgan Ames1 Mor Naaman
Department of Communication Yahoo! Research Berkeley
Stanford University Berkeley, California
morganya [Link] mor [Link]
ABSTRACT is driven by the proliferation of capture devices (such as
Why do people tag? Users have mostly avoided annotating digital cameras and cameraphones) together with
media such as photos – both in desktop and mobile decreasing storage costs. At the same time, creation of
environments – despite the many potential uses for semantic metadata about photo content remains an elusive
annotations, including recall and retrieval. We investigate goal. Some amount of annotation can significantly improve
the incentives for annotation in Flickr, a popular web-based the usefulness of such photo collections as they grow into
photo-sharing system, and ZoneTag, a cameraphone photo the thousands; the most commonly-cited benefits are to help
capture and annotation tool that uploads images to Flickr. In recall and support search [11]. However, algorithms for
Flickr, annotation (as textual tags) serves both personal and semantic interpretation and annotation of image content are
social purposes, increasing incentives for tagging and far from reach by any automated system. As a result, the
resulting in a relatively high number of annotations. owner of the collection is tasked with the labor of
ZoneTag, in turn, makes it easier to tag cameraphone annotating their photographs.1
photos that are uploaded to Flickr by allowing annotation
Providing tools for annotation of media is therefore an
and suggesting relevant tags immediately after capture.
active field of research in human-computer interaction [3, 8,
A qualitative study of ZoneTag/Flickr users exposed 21]. These research efforts, mainly focused on desktop-
various tagging patterns and emerging motivations for based tools, attempt to ease the annotation task as well as
photo annotation. We offer a taxonomy of motivations for maximize the benefits for annotating content [10]. While
annotation in this system along two dimensions (sociality some of this research has been incorporated into
and function), and explore the various factors that people commercial photo browsing systems (like Adobe Photoshop
consider when tagging their photos. Our findings suggest Album), most people still do not bother with annotation of
implications for the design of digital photo organization and their photos [17], though they do say they think it would be
sharing applications, as well as other applications that useful for photo retrieval and wished they did it much more.
incorporate user-based annotation. Clearly, the perceived benefits of annotation – which are
vaguely-defined and at some indeterminate time in the
Author Keywords future – do not overcome the investment [7], even with the
Annotation, tagging, digital photographs, cameraphone, most advanced annotation systems.
incentives, motivations, photo sharing, collection
organization, location-aware, CSCW. The motivations for tagging changed with online photo
sharing communities such as Flickr [1]. Flickr allows
ACM Classification Keywords annotation of photos in the form of tags, or unstructured
H.5.m. Information interfaces and presentation (e.g., HCI): textual labels. The tags in Flickr are mostly assigned by the
Miscellaneous. user who uploads the image [11], and provide multiple
benefits. In addition to making the photo searchable by the
INTRODUCTION contributing user, tags enable users to discover other users’
Innovations in consumer photography have made it photos. In other words, the traditional use for annotation,
exceedingly simple for people to capture images, which personal organization and retrieval, is now augmented by
they do at an ever-growing rate. This growing capture rate the ability for users to expose their photos on-line to be
viewed by other members of the Flickr community.
ZoneTag is a cameraphone application used to upload
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are photos taken by the phone to Flickr. ZoneTag is loosely
not made or distributed for profit or commercial advantage and that copies based on the MMM system [19] designed to capture,
bear this notice and the full citation on the first page. To copy otherwise, annotate, store and share photos from the phone. Most
or republish, to post on servers or to redistribute to lists, requires prior
specific permission and/or a fee.
CHI 2007, April 28–May 3, 2007, San Jose, California, USA. 1
Copyright 2007 ACM 978-1-59593-593-9/07/0004...$5.00. This research was conducted while the first author was at Yahoo!
Research Berkeley.
importantly, ZoneTag attempts to encourage annotation on are designed into a game. ESP randomly matches two
the phone at the time of capture by providing tag physically and virtually separate users (the two cannot
suggestions. communicate). The game simultaneously presents the users
The combination of Flickr and ZoneTag allows us to study with the same photo and users earn points if they type in the
this new class of photo organization, annotation and sharing same textual label.
environments. Indirectly, the study reflects on a broader Collaborative tagging and annotation systems, for photos as
class of web-based “tagging systems” that have recently well as other resources, are a growing area of interest in the
become commonplace [11]. The motivations and incentives HCI community. Golder et al. [4] studied tagging dynamics
for participation and tagging in these systems have not been and reported on usage of different types of tags in
widely studied to date. We attempt to find out why users of [Link] ([Link] Sen et al. described personal
Flickr “tag” their content. The study may shed light on and social influences on tagging behavior in MovieLens
motivations in other systems too, a direction we would like [20]. Marshall [12] analyzed annotations in reading material
to pursue in future work. along a number of dimensions including formality,
Our contributions focus on the tagging activity in ZoneTag permanence, and explicitness, among others. Though her
(and Flickr). Specifically, we: work is focused on personal annotation, the informal
“ecological” annotations she observed are similar to the
• Examine the motivations for annotation and tagging in tagging in ZoneTag and other social tagging systems.
Flickr and ZoneTag via user interviews and other Finally, Marlow et al. [11] offer a taxonomy of
qualitative methods. collaborative tagging systems; report on different incentives
• Describe a taxonomy of motivations to annotate content for tagging, categorized as social vs. organizational; and
in this system. provide an initial study of the tagging dynamics on Flickr.
• Study the role of tag suggestion in the system; We extend their work by providing a more detailed
specifically, the ways these suggestions affect the taxonomy of tagging motivations on Flickr. In addition, we
behavior and patterns of tag use. ground our work in user interviews and other qualitative
We offer implications for designers of systems for photo methods, providing a deeper understanding of real-world
annotation and organization; for designers of other tagging user attitudes.
and annotation systems; and for designers of mobile capture The idea of sharing media metadata between users to assist
devices like digital cameras and cameraphones. in annotation was introduced in earlier work [13, 19]. In
this form of collaborative tagging, metadata such as tags
RELATED WORK and labels is shared, mostly based on spatial and temporal
Efficient labeling of photos has been an active research context. The system developed in [19] for cameraphones
field since 1999. Work in [3, 5, 8, 21, 25] addressed ease was the first iteration on a collaborative tagging client for
and partial automation of the labeling task on one hand, and the mobile phone platform. That work proposed annotation
expanding the benefits of annotation on the other. For using person, location, object and activity categories. As we
example, [21] proposed a drag-and-drop approach for show below, a number of key conceptual and usability
attaching labels to identify people in photos. The latest lessons from that system and subsequent efforts were
photo browser commercial packages, such as Adobe incorporated into ZoneTag.
Photoshop Album, adopted similar methods to support easy
The HCI community has recently become interested in
labeling of photos. More recent efforts [3, 14] utilize
investigating current cameraphone usage [6, 15, 22],
temporal and spatial context to assist in labeling photos in
developing new experiences around mobile photos [18],
personal collections. However, a recent study by Kirk et al.
and studying new forms of sharing [22]. Although the types
[7] on management of personal photo collections has
of pictures taken in ZoneTag show patterns similar to the
determined that users often use only event-based
ones described in [6, 22], it is the first such system that uses
organization for their photos and apply little or no
a popular, web-based photo sharing system such as Flickr,
additional annotation or organization.
with an audience much wider than those actually using
Some projects have considered collaborative work on ZoneTag. In addition, Flickr differs from traditional photo-
labeling photographs, allowing many users to label a shared sharing sites in that photos are by default public, enabling
collection of images. In [9], participants annotated a public one-to-many sharing rather than one-to-one. We describe
collection of photos from the CHI 2001 conference. While Flickr and ZoneTag next.
this study is geared towards an existing community, a
broader example can be found in current online photo FLICKR
sharing systems like Flickr [1]. In Flickr, users label photos Flickr is an active photo sharing web site and community
with tags or captions. Communities and photo pools are [1]. At the time of writing, Flickr has over 5 million
often created in an ad-hoc manner as users annotate with registered users and over 250 million images. There are
the same tags (e.g., people interested in photos with the various controls and settings associated with each photo
“CHI 2007” tag). The ESP game [24] offers a different uploaded to Flickr; the relevant ones are listed here. First,
version of collaborative tagging, where the user incentives privacy settings allow users to specify whether each photo
 ZONETAG
This section describes the key relevant aspects of the
ZoneTag application that iterates and improves on previous
efforts [19]. In particular, we briefly discuss the photo
capture and upload flow, and expand on the post-capture
interface that allows the user to associate metadata in the
form of tags to the photos.
ZoneTag is a mobile application available as a public
prototype for Nokia and Motorola phones.3 Using ZoneTag,
users can upload a newly-captured image from their
cameraphone to Flickr [1] in as few as two clicks. After a
photo is captured, ZoneTag displays an upload dialog over
the image (see Figure 2). The user can review the image
Figure 1. A ZoneTag photo page on Flickr; tags are displayed and decide whether she wishes to upload it to Flickr. In
on the right. Visitors to the page can click on a tag, or on the such cases, more options are shown. Clicking the same
earth icon next to it, to see photos with this tag from this user button again will upload the image, keeping all the settings
or from the entire Flickr community, respectively. and tags entered for the previous photo.
can be viewed by the public, by the users that Flickr Most importantly, the user has an option to type in or select
members have specified as friends and family, or just by the tags that will appear when the photo is displayed on Flickr.
user. Second, the user can assign a title and a caption to The tag-entry screen includes context-based tag suggestions
each photo (placed above and below the photo, that are pre-fetched from the ZoneTag server and sorted by
respectively). the likelihood that the user will select a particular tag, given
In addition to these settings, the user can enter a set of contextual information described below. The tags are
textual labels for each photo, also known as tags.2 The grouped into categories reflecting tag sources. The tag
Flickr interface displays these user-entered tags to the right categories, shown as tabs in the user interface (see the right
of the photo on the photo’s page (Figure 1). Thus, one side of Figure 2), include ‘local tags,’ which are tags
function of the tags is descriptive, providing additional created by the user, by members of their social network, or
information about the photo that may not be reflected in the by others in the current location; ‘recent tags,’ which are
title or the caption. tags the user has used within 24 hours; and a
comprehensive list of ‘all tags,’ including tags the user has
Users can also use tags to search for photos in their own
either entered on the phone or applied to a ZoneTag photo
collection, in another user’s collections, or across the entire
on the Flickr website. In addition to user-contributed
public pool of photos. A search mechanism returns photos
contextual tags, ZoneTag incorporates tags from sources
whose title, caption, or tags matches the search query. A
like place and event databases. These ‘canned’ tags are
tag-specific retrieval mechanism is also available: a user
served and sorted based on physical location. Within each
can search for all photos with a specific tag in their private
tab of tags, a user can quickly search through the available
collection, in another user’s collection, or across all public
tags by scrolling or by entering the first few letters of a tag
photos.
in the search box.
Optional pre-upload settings allow users to control various
features, including the photo’s privacy settings, the title for
the photo, and the presence of ZoneTag’s automatic
location tags (such as city name).
The details of how tag lists are generated, sorted, and
transmitted to the mobile device exceed the scope of this
paper. Briefly, ZoneTag considers location and time as
primary sources of context. The system aggregates all tags
from a similar context, and then ranks them by frequency
and various likelihood measures, giving more weight to the
user’s tags and then to tags used by people in the user’s
social network. For example, if a user takes photos at home
Figure 2. The upload dialog (left) and the tag suggestions and tags them with “home” and the names of others present,
and entry screen (right) in the ZoneTag interface. all these tags will appear as suggested tags when the user
takes additional photos at that location. In another example,
2
A user can allow other Flickr users to add tags to his or her
photos. However, the practice of adding tags to photos taken by
3
others is not prevalent [11] and we shall ignore it here. Our description here applies to the Nokia version of ZoneTag.
when a user takes a photo at a location where many users The relative number of users for different ranges
have used the tag ‘Golden Gate Bridge’, that tag is likely to of tag usage (n=172)
be suggested. 30%

We found that ZoneTag was especially interesting as a 25%

platform for the study of emerging tagging motivations 20%
because it has incorporated some significant lessons from
previous systems, including the following. 15%

• ZoneTag uses a vibrant photo sharing platform, Flickr, as 10%

a destination for the generated content. In Flickr, photos 5%
are guaranteed to be archived for a long period, creating
further motivation for photo capture. In addition, using 0%
<0.5 <1 <1.5 <2 <2.5 <3 <3.5 <4 <4.5 <5 5+
Flickr, ZoneTag users can interact with users outside the
ZoneTag system, thus potentially expanding the sharing
and interaction around their photos to millions of users Figure 3. ZoneTag users’ tagging frequency across their
who are not in the ZoneTag system. entire Flickr collections (including untagged photos).

• ZoneTag provides various benefits for tagging photos, short analysis below are to examine whether the data
mostly based on Flickr’s affordances for tags. In past indicate emerging trends in how ZoneTag users use tags for
work, there was no immediate value to users annotating their photos, and whether the tag suggestions have some
their photos. effect on the usage of the system.
• In previous systems, the user had to annotate each photo The collected data included the settings applied to images
before it was uploaded. In ZoneTag, we made the through ZoneTag and on Flickr including privacy settings,
annotation step optional on the path to upload. Users who image title, and tags, allowing us to closely inspect the
are not interested in tagging are not burdened by the extra user’s tagging activity. For example, we could learn which
step and still find ZoneTag useful. Moreover, when tags were added to each photo on the phone and which were
necessary, even users who would usually tag their photos added on the web using the Flickr interface. Moreover, for
can upload a new photo quickly with no tags, or with the tags added from the phone, we can tell whether the tag was
previous photo’s tags still attached. typed in or selected from the list of ZoneTag suggestions.
• The tagging interaction itself is improved significantly, Figure 3 shows the percentage of ZoneTag users (y-axis)
optimized for simplicity and speed. Tag suggestions are whose average number of tags fell into each range (x-axis).
pre-fetched from the server and shown without delay We include all tags that users added to their photos,
when the user chooses to view the tagging dialog. In whether on the phone or on Flickr, as well as photographs
previous systems, tags were not pre-fetched and the in the users’ ZoneTag collections with no tags. For
tagging dialog only appeared after a long delay, which example, the leftmost bar indicates that for 25% of the
made the system considerably less usable. ZoneTag users, the average number of tags per photo in
• Previous systems solicited annotations in multiple their collection was smaller than 0.5 (in other words, they
semantic categories (e.g., “place”, “event”, “activity” and added one tag to every other photo or fewer, on average).
so forth). In our system, the annotation – in particular, the The aggregate of all bars, excluding the leftmost two,
tags and the tag suggestions – are drawn from a flat un- indicates that 61% of ZoneTag users added, on average, at
categorized space, which seems to be more conceptually least one tag per photo. Note that this count does not
accessible to users than faceted metadata. include the automatic tags that ZoneTag adds to photos,
such as location names and other default tags.
DEPLOYMENT Our users’ approaches to tagging on the mobile phone or on
ZoneTag has been deployed as a publicly-available Flickr varied considerably. Of the users that regularly tag
prototype for over nine months. Most, but not all of the their photos, many do so on the phone as they upload a
users of ZoneTag are self-selected early adopters of picture, using the ZoneTag interface. Over two-thirds of the
technology (we found later that similar patterns of tagging ZoneTag photographs that were tagged at all were tagged
and contribution arose within both technical and non- from the phone. Users also may revisit their photos on the
technical taggers). As of January 2007, ZoneTag was used Flickr website to add tags. Some users exclusively choose
by more than 500 people that uploaded over 45,000 photos this web-based route for tagging, never adding any tags on
to Flickr, an average of about 90 photos per user. At the the phone.
time of the study, 172 users had taken more than 10 photos
using ZoneTag; our data analysis below mostly focuses on Initial evidence demonstrates that the tag suggestions in
these users. ZoneTag have increased the number of tags associated with
photos in the Flickr interface. The users of Shozu
During deployment, we have collected detailed data ([Link] another cameraphone application that
regarding the usage of the system, which we have used to uploads images to Flickr, provide a point of comparison.
examine tagging patterns and activity. The goals of the
The installation, distribution, usage, and supported phones look at the different motivations for tagging as found in the
of Shozu are all similar to ZoneTag. As a result, we assume interviews. Finally, we report on how tag suggestions
relatively similar user characteristics between the two influenced our participants’ tagging behavior.
systems. A key difference between the systems is that in
Shozu, if users wish to tag their images on the phone, they Reasons for Photo Capture
are required to type the tags in without the application’s aid. Generally, the types of photos captured in our system did
Even though users in both systems could always add not stray from previous research findings for photos in
missing tags on the web using the Flickr interface, the cameraphones and digital cameras [2, 6, 15, 22]. For
easier tagging mechanisms in ZoneTag result in a larger example, Kindberg at al [6] have classified the reasons for
number of tags per photo overall (including web-entered cameraphone image capture into two dimensions: social vs.
tags). Based on a recent sample of Shozu images on Flickr, personal, and affective vs. functional. In particular,
the average number of user-entered Flickr tags for a public social/affective photographs are photos taken for sharing
Shozu photo is 0.97 (standard deviation = 2.05, n = 4087). with friends and family members (either to share mutual
The number of such tags for a ZoneTag public photo is 2.2 experience or to connect with absent friends and family).
(standard deviation 2.15, n = 18417). Despite the high Indeed, despite the one-to-many sharing aspect in ZoneTag
variance, these numbers may indicate some effect of and Flickr, the usage of ZoneTag and Flickr amongst all our
ZoneTag’s tagging environment. participants (and seemingly amongst the broader ZoneTag
user base) is similar to the usage of mobile photos as
The patterns that exist in ZoneTag and Flickr data have reported in [6], with one significant difference. In our
indicated that while users’ tagging behavior varies, system, the fact that the photos are uploaded to a
ZoneTag appears to aid in tagging. These patterns also (potentially) publicly-accessible web site can introduce
informed us that many different factors and motivations additional motivations for taking and sharing photographs.
come into play for users who tag their photos. For example, These motivations are best described as artistic exposure
the average number of tags attached to a non-public and recognition.4 Using the taxonomy of Kindberg et al,
ZoneTag photo is 1.85 (standard deviation 1.892, n = this type of usage would still fall under the social and
24089), lower than the average of 2.2 tags attached to a affective usage categories. However, in this case, the photos
public ZoneTag photo (again we point out the high are sometimes intended for general public as well as social
variance). In any case, we set out to investigate these cohorts. As we show below, this new category of sharing
motivations qualitatively via open-ended user interviews. motivations also drives some of the tagging activity in
INTERVIEWS Flickr and ZoneTag.
The main part of our study involved in-depth, semi- The relatively high image quality available from the
structured interviews with 13 participants, including some cameraphones used by our participants – especially
of the ZoneTag users who had taken the most photos. Four compared to the image quality of cameraphones used in
of our participants were female. The participants ranged in previous studies – may have also influenced their reasons
age from 25 to 45 with most in the 25-35 range. While there for photo-taking (most participants’ cameraphones
was an over-representation of technical people (9 out of supported a resolution of 1.3 to 2 megapixels). We suspect
13), we found that the patterns of use did not differ much that a higher number of photos were taken for archival-
between our technical and non-technical participants. related purposes than the numbers described in [6, 22].
Amongst our participants, the average number of user- While we did not try to reproduce the numbers reported in
added tags per participant per photo ranged from zero to these studies, there is certainly a sense of moving from
more than five. capturing throwaway images with the cameraphone to more
Our interviews consisted of a casual conversation about regular, digital-camera-like capture behavior. For instance,
Flickr and ZoneTag use, followed by a more detailed three of our interview participants reported taking their
discussion of tag usage and motivations for image capture ZoneTag cameraphones, and no other camera, on vacation
and tagging. In a final, directed portion of the interview, we with them.
used the photo elicitation technique discussed in [23] to ask To summarize, the users we interviewed and the photos
about specific tags used by participants for their they took have similar characteristics both to mobile
Flickr/ZoneTag photos. We viewed Flickr photos and their sharing behavior (always-there capture device, ease of
associated tags with the participants and discussed specific sharing) and traditional photo taking (better quality,
patterns in tag activity such as sudden spikes and drops. archival purposes). Our observations are therefore
Finally, we asked the participants about tag suggestions and potentially applicable to a wide range of photo sharing
their understanding of the source of these suggestions. applications and to different capture devices, including both
digital cameras and cameraphones.
RESULTS
We first discuss key issues regarding the type of photos
taken by our participants and the motivations for photo 4
Van House et al. [22] found similar artistic expression activity in
capture. Once the reasons for capture are established, we one-to-one sharing of cameraphone photos.
Motivations for Tagging Self/Organization: Search and Retrieval
Our interviews surfaced a wider variety of motivations and The “self/organization” set of motivations, in the upper left
uses for annotation and tags than discussed in previous corner of Table 1, represents the traditional annotation
studies. Few participants annotated their ZoneTag/Flickr motivations in personal photo collections [3, 21].
photos solely for the traditional organization and self- Participants that are driven by this motivation to tag
retrieval benefits. Moreover, there were often multiple sometimes made comments such as “I am an organized
motivations involved even in the use of a specific tag for a person” or “I like order.” Often these comments are
specific photo. Still, we found that most of our participants accompanied by apologies for not being more exhaustive or
generally had one or two primary motivations for tagging. consistent in tagging, or by guilty laughs for being
For instance, one participant visited Hawaii and tagged his “obsessive” about it. Two participants (see Quote 1 for one
photos with tags that both intended to help others find his example) say they are motivated to tag specifically to later
photos, and conveyed information and opinions about the retrieve their pictures for sharing; two other participants
photo itself, such as “Aloha Air sucks” or “good also report tagging for personal organization purposes.
restaurant.” Another participant added tags to allow herself
to find her photos later, as well as to provide contextual Quote 1 (P6). Mostly I use [tags] if I go back on Flickr, if I
information for friends and family viewing her photos. want to find all the pictures of one thing. If I tagged ahead of
time I can go back and get all my pictures of [my children]. …
We developed a taxonomy for the revealed set of I’ve made separate tags of [my child’s] preschool or playgroup
annotation motivations, as shown in Table 1. There are two so that if I want to share pictures with more than just family I
dimensions along which we place the different incentives can go back and find everything from that one tag. … Mostly
for tagging photographs. The first dimension, “sociality,” it’s for my own organization at this point.
relates to whether the tag’s intended usage is by the Quote 1 illustrates one participant’s self/organizational
individual who took and uploaded the photo or by others, motivations for tagging. The participant code (e.g. P6 is
including friends/family and strangers. This is similar to participant 6) is in parentheses.
Kindberg et al’s. The second dimension, “function,” refers
to a tag’s intended uses. We found that users tagged their
Self/Communication: Memory and Context
pictures either to facilitate later organization and retrieval or
Sometimes participants enter tags to add context to a
to communicate some additional context to viewers of the
photograph, such as the names of the people that appear in
image (whether themselves or others). Our “function”
it or the name of the place it was taken, in order to aid
dimension focuses on the motivation for adding tags and
future recall of the situation it depicts (see Quote 2 for an
differs from Kindberg et al’s functional/affective
example). The “self/communication” section of our
dimension, which focuses on the intended use of the
taxonomy (the top right square in Table 1) reflects this set
photograph itself and does not differentiate communication.
of motivations.
The social category of the “sociality” dimension (second
Interestingly, though adding context is the traditional
row of Table 1) is where tagging incentives that are not part
motivation for annotating printed photographs [16],
of traditional personal media management are introduced.
relatively few participants were motivated in this way when
These ‘social’ motivations, as discussed later, are actually
tagging their ZoneTag/Flickr photographs. Even when users
by far the most common motivation for tagging that our
made their cameraphones their primary capture device, and
participants reported. As more people use their
despite the fact that their cameraphones produced pictures
cameraphones and upload their images, we expect that
that are good enough for archival purposes, we found that
these social motivations will continue to grow.
the memory function of tags was still not a popular
In the following sections, we discuss each of the motivation motivation. While it is likely that tags will provide this
categories listed in Table 1 and provide examples from our function as a currently-unanticipated benefit in the future,
interviews. We then report the motivations that are primary adding context to facilitate remembering details about
for each of our users. photographs was not a primary motivation for tagging in
the present.

Quote 2 (P2). If I have the time, the neighborhood, or the

event, I have enough information to look at my own collection
and know where this came from. I don’t have the bandwidth to
tag for the benefit of the Flickr system. … I want at least one
hook of association in there that can help me reconstruct what
I was thinking. I don’t have time to put all the hooks in but I
can put one in.
Quote 2 illustrates one participant’s self/communicative
motivations for tagging.
Table 1. A taxonomy of tagging motivations in ZoneTag/Flickr.
Social/Organization: Public Search and Photo Pools As more people, including groups of friends, tag their
The “social/organization” category of our taxonomy (the photographs, other interesting behaviors emerge. Two of
lower left square in Table 1) represents the user’s our participants reported coordinating tags with others in
motivations for making their photos findable by others. The order to facilitate later search and retrieval, forming, in
reasons for capturing the corresponding photos are social as effect, an ad-hoc, distributed photo “pool.” One participant
well, and mostly “affective” rather than functional. As did this explicitly with other users in many settings, ranging
Kindberg et al. point out [6], photos are often captured to from company meetings to parties, from classrooms to
enrich or document mutual experience, or to share hikes with friends (Quote 6). Another attended a race and
experiences with remote friends and family; in Flickr, parade in San Francisco and used the tag that others were
photos can also be shared with the public. The tagging using in order to tag his own photos and to find others’
motivations that follow ensure that the photos can be easily photos of the event later. Still another heeded his friends’
found by specific people a user might want to share with wishes in adding their usernames to photos with them in
(such as friends of family), or discovered by anyone who them so that they could find pictures of themselves later
may be interested in the photo. (Quote 7).
Flickr is fairly unique in having a large and vibrant public Participants also reported being motivated to include tags so
photo pool. While some participants tagged specifically for that the tags would be suggested to other users taking
their family or friends (e.g. Quote 5), others had this wider photos in the same location. At least three interview
audience in mind when adding tags (e.g. Quote 3). One participants explicitly mentioned tagging their photos in
participant said that friends and family followed his photos part to make the tags appear in others’ suggested tags list
and didn’t need to use his tags to find photos; thus, his tags (see Quote 11 for an example). While these users did not
were just for the general public. explicitly coordinate their tagging with others, they realized
While seemingly altruistic, even sharing photos for the this “coordination” would happen implicitly. If successful,
general public has, in the Flickr system, personal this implicit coordination will allow the users to search for
motivations. A clever feedback system alerts the user to the the photos others took of the same event or in the same
number of times each photo was viewed, chosen as a place. This motivation was facilitated by participants’
favorite by others, or commented on, providing the knowledge of the suggested tags feature of ZoneTag, which
photographer with the satisfaction of knowing their pictures in effect helps to standardize the tag pool in Flickr. We
are getting attention and that they are gaining a good provide more discussion on suggested tags below.
reputation in the Flickr community.
Quote 6 (P8). I’m at an event and there’s a convergence on a
Two participants described ways of “gaming” the system by specific tag, then I’ll tag because it’s for the good of the group.
using certain tags to attract more views (e.g., Quote 4). In … It’s a nice way to build live streams and collections of
some cases, the participant also wants to promote the photos. … [A classmate] suggested we tag everything
subject of the picture (for one participant, a local band) as specifically so we can find it, which is actually really useful.
well as the picture itself. We and others are studying Quote 7 (P3). If I’m out with friends … they might suggest
additional motivations for posting pictures publicly on sites tags. A couple of my friends will say “put my user ID in there”
such as Flickr, but these are beyond the scope of this paper. so they can find the picture. ... Using my user ID on my Flickr
photos pulls my photo into certain streams.
Quote 3 (P6). I tagged [this new restaurant] because it just Quote 6 and Quote 7 illustrate another aspect of
opened and it was new and I wanted to tell people what it was. social/organizational motivations for tagging: photo pools
… It was a new place and probably that would be useful if and tag normalization. Both quotes are examples of tagging
other people were going there for the first time. primarily for friends and family.
Quote 4 (P3). Mostly friends view my photos, but as I grow Social/Communication: Context and Signaling
my collection, I am getting more [public] views. I've noticed
The final category in our taxonomy (in the lower right
that if I take [and tag] pictures of cute female friends, views go
up. … There’s a satisfaction that 50 people have viewed my corner of Table 1) includes the “social/communication”
photos. I know that tagging can connect my photos to motivations for tagging. Here, participants tag in order to
activities, and get more interest. … I'm getting more liberal communicate contextual information to others about the
about using suggested tags lately, so will add multiple tags to image and consequently about themselves as
make it easier for people to find my photos. photographers, as Marlow et al. discuss in more detail [11].
Quote 5 (P6). When my sister’s baby was born she got a new In most cases, participants added these contextual tags for
tag, and at one point my dad was clicking on the [child’s the benefit of known others, such as friends or family – see
name] tag … [the tag provided a way to] sort by child. Quote 8. Only one participant indicated adding descriptive
Quote 3 and Quote 4 illustrate one aspect of tags for the general public: in Quote 9 the participant
social/organizational motivations for tagging: public search describes leaving restaurant and airline reviews in his tags.
and self-promotion. Quote 5 is an example of tagging so that Adding descriptive tags for known others often involved
friends and family can search; the rest focus on the general
using tags that have little meaning without additional
public.
context, as is the case with inside jokes or nicknames
(relating to the discussion of “common ground” in [6]). In
fact, some participants, in order to maintain privacy,
purposefully obfuscated their tags so that friends or family
would understand what the tags meant but the general
public would not.
The two participants who had several friends and
colleagues using ZoneTag reported an additional in-the-
moment social/communicative motivation for taking
Table 2. Primary (bold) and secondary (italic) motivations for
pictures. They found that taking and tagging was a socially
tagging among our 13 interview participants.
contagious activity. If one person took a picture, others also
took out their cameraphone: the act of taking a picture
2. While we did not have enough participants to indicate
“signaled” the event as photo-worthy, as also observed by
significant trends, we can suggest that most of our
Van House et al. Additionally, these two participants
participants were motivated to tag by organization for the
reported using tags as a way of exchanging information and
general public (photo pools, search, self-promotion), with
inside jokes with friends in the moment, exploiting the self-organization (adding tags for later retrieval) and social
socially- and locally-situated tag-sharing feature of communication (adding context for friends, family, and the
ZoneTag. One described this group photo-taking and
public) tied for second. The predominance of social
tagging as “a chain reaction”: “someone takes out their
motivations for tagging coupled with the success of tagging
camera, then others take out their camera thinking
on Flickr suggests that adding this social dimension to
something important is happening.” The other described
tagging greatly increases the likelihood that users will tag
such photo-taking as a “shared social experience” (see
their photos.
Quote 10).
In Table 2, we note an interesting zigzag pattern of
motivations. First, organization for oneself is a more
Quote 8 (P4). [I tag] so I don’t have to explain myself – so my common motivation than communication for oneself. While
friends don’t have to ask me a billion questions about ‘where did just over half of our participants used tags for personal
you take this photo, why are you showing me this photo, who is search and retrieval (a motivation focused on in many
this person in this photo’…I can give them the basic story. previous papers on tagging), very few were motivated to tag
Quote 9 (P11). I left reviews of places – like at the airport, when for personal memory. However, we expect that this will be
my flight was delayed, I tagged “Aloha Air sucks.” an unanticipated future benefit, even though it has not been
Quote 10 (P7). It’s like digital [yawning] or something. One an initial motivation for most participants. For instance,
person pulls their [camera]phone out, and then everyone starts when these photos are viewed years from now, the tags can
pulling out their camera. My use [of ZoneTag] went way up provide additional information and jog memories much the
when my team started using it. … When I'm with other friends same way that captions and descriptions written on the
who don't have ZoneTag, there’s no point in using tags like that. backs of prints do today.
The social dynamic isn’t as fun. … When there is a social
dynamic, my photo taking goes way up. It’s a shared social Second, communication with friends and family is a more
experience. common motivation than organization for friends and
Quote 8 and Quote 9 illustrate social/ communicative
family (six participants vs. two). Very few participants were
motivations for tagging. Quote 10 illustrates one participant’s motivated to tag to help friends and family find their
in-the-moment motivations for picture-taking and tagging, which photographs; most said they would either send photos
occurred with other friends who use ZoneTag. directly or found that their friends and family kept track of
their photo collection as it evolved and didn’t need to use
Summary of Motivations tags to search for photos. However, like personal memory
Again, we emphasize that specific tags can play several described above, this may be a future benefit that is not a
roles in our motivation taxonomy. For example, a place- motivation in the present.
name tag can be used both for potential retrieval by the user Finally, organization for the general public is a much more
and by others, for communicating location to others, and for common motivation than communication. Only one
reminding the user where a picture was taken at a later date. participant was motivated to add contextual tags aimed at
However, we found that generally, most participants only the general public (see Quote 9), and only in a few
considered one or two motivations for adding tags; in many situations (giving restaurant and airline reviews), while 10
cases, they had not considered the other possible benefits. of the 13 participants added tags to make their photos
With this in mind, we analyzed the interviews again, trying publicly findable. It appears that the motivations for adding
to extract the possible main motivating categories for each contextual tags, which tended to be very personal, are
of the interview participants. The results are shown in Table largely unrelated to the motivations for adding tags for the
general public, namely, to make one’s photos findable and
to gain reputation in the Flickr community.
Suggested Tags purpose that simply aiding in tag entry. First, some
Part of our interview focused on the ways our participants participants had become used to scanning the tag
understood and used ZoneTag’s suggested tags. What was suggestions to add any tags that may be relevant, even if
the effect of the tag suggestions on the users’ tagging they did not mean to add them in the first place. One
behavior and activity? As we discovered, tag suggestions participant described how he scrolls down the list and
had various implications; but first, we must note the selects available tags simply because they are present. Less
importance of the ZoneTag feature allowing users to skip helpfully, some participants stated they add such tags even
tag entry on the phone. if they are not entirely relevant or accurate.
As noted above, an early design decision in ZoneTag was to Second, the suggested tags, even when not selected,
make the interaction before photo upload as burden-free as inspired some participants to add their own tags and gave
possible. In particular, ZoneTag made it possible to upload them direction as to the sorts of tags they should use (e.g.,
a new photo in two clicks without even looking at the seeing other neighborhood names as suggested tags could
settings, let alone adding or selecting tags. This 2-click encourage the user to add a tag for their neighborhood as
upload was a welcome feature. First, many users found it well).
difficult to enter or even select tags on the phone, especially
To summarize, ZoneTag’s tag suggestions have had a large
in certain situations such as while driving or socializing.
impact on the participants’ tagging activity. However, the
Even participants who did add tags on the phone did not
option to bypass tags altogether was an important feature
always want to be interrupted in the moment of capture.
for the usability of the system as a whole.
Many participants reported setting up tags once for an event
and then uploading photos with the same tags without IMPLICATIONS
having to interact with the tagging interface after each Our observations and analysis of user motivations in the
photo. In Quote 12, the participant liked that previously- Flickr/ZoneTag system allow us to draw implications for
used tags showed up (though a few participants reported the design of photo sharing, annotation, and organization
accidentally adding tags they didn’t intend to add because applications, as well as for tagging systems in general.
of this feature). In Quote 11 the participant modifies the • Make the annotation pervasive and multi-functional, and
way he tags, knowing that others will see his tags as incorporate motivations in all four categories of the
suggested tags when they are in the area. Many participants taxonomy. For example, tags or captions created by the
also used the auto-completion feature extensively (e.g. see user for a photo on a desktop photo application should be
Quote 12). displayed, and made searchable, in web albums where the
The function of the suggested tags was clear for most, but photo is published, or in email messages sent with it.
not all, of our users. Some participants commented about • Make it easy to annotate when the information (photos, in
the quality of the suggestions. One said that ZoneTag this case) is captured. The higher rates of annotation in
worked great when she was attending a conference with ZoneTag than in Shozu (where it is more difficult to tag)
other ZoneTag users who created local tags: “I thought, or Flickr (where tagging takes place after-the-fact) show
great! This is how I want it to work all the time!” that easy annotation at the point of capture seems to
On the other hand, we found that tag suggestions sometimes increase tagging activity (also see [22]).
did not work well. In particular, since the system cannot • However, do not force users to annotate at the point of
distinguish different types of tags, unknown person names information capture. The system will become unusable
entered by another user near the same location sometimes for people who would not annotate [13]. In mobile
appeared in the tag list, thus confusing some participants. conditions, even users who are inclined to annotate would
One participant commented about an unfamiliar name, not always be able to.
“This person was in my phone for a month! Who is she?” • For systems that have both mobile capture and
We also learned that suggested tags served a broader desktop/web based components, allow annotation in both
settings: leverage the fast entry and powerful tools of the
desktop to allow more descriptive or bulk annotation, and
Quote 11 (P13). I was taking a picture of the water tower in
Sunnyvale, and I thought about how my tags would show up
the in-the-moment aspect of mobile devices to remind
for others. … Tag suggestions were huge for me; they really users that they could take a moment to annotate now.
cut down on typing. • Relevant tag suggestions, even when not used directly,
Quote 12 (P2). I appreciate the fact that I can reuse the tags. It can encourage tagging and give users ideas about
makes it worth my while. … I try to use as many suggested possible tags. However, suggestions should be used with
tags that apply. … I also use it for auto-completion – I type caution. First, users may be confused or alarmed by
“s” to get San Francisco. inexplicable tags. Second, users may just choose these
Quote 11 and Quote 12 illustrate the usefulness of suggested tags even if they are not immediately relevant to the
tags. All participants reported liking the system, even when content instead of manually entering more accurate tags.
they didn’t fully understand it.
CONCLUSIONS 5. Kang, H., and Shneiderman, B. Exploring Personal Media: A
The topic of content annotation has been an important Spatial Interface Supporting User-defined Semantic Regions.
research area in the field of human computer interaction. Tech. Report ISR 2005-51, U. Maryland, 2005.
We conducted qualitative studies in a real-world system 6. Kindberg, T., Spasojevic, M., Fleck, R. and Sellen, A. The
where users annotate their data – in this case, photos. The Ubiquitous Camera: An In-Depth Study of Camera Phone Use. In
interviews, although focused on a narrow set of users, IEEE Pervasive Computing 4, 2 (2005), 42-50.
showed that more than one set of motivations comes into 7. Kirk, D.S., Sellen, A.J., Rother, C., and Wood, K.R.
play when users tag their photos. We hypothesize that Understanding Photowork. In Proc. CHI ’06, ACM Press (2006).
having these multiple motivations is a determining factor in 8. Kuchinsky, A., Pering, C., Creech, M.L., Freeze, D., Serra, B.,
users’ decision to annotate. In particular, social incentives and Gwizdka, J. Fotofile: A Consumer Multimedia Organization
and Retrieval System. In Proc. CHI ’99, ACM Press (1999).
for tagging appear to be surprisingly important in
motivating users to tag their photographs. 9. Kules, B., Kang, H., Plaisant, C., Rose, A., and Shneiderman, B.
Immediate Usability: Kiosk Design Principles from the CHI 2001
Under these conditions, we showed that it is possible to Photo Library. Tech. Report CS-TR-4293, U. Maryland, 2003.
motivate users to annotate content. Point-of-capture 10. Kustanowitz, J. and Shneiderman, B. Motivating Annotation for
annotation (e.g., on the mobile device) can encourage the Personal Digital Photo Libraries: Lowering Barriers while Raising
addition of tags. Tag suggestions and other methods of Incentives. Tech. Report HCIL-2004-18, U. Maryland, 2005.
assisting mobile annotation proved to have broader 11. Marlow, C., Naaman, M., boyd, d., and Davis, M. HT06, Tagging
implications that just assistance in text-entry. In some cases, Paper, Taxonomy, Flickr, Academic Article, ToRead. In Proc.
the suggestions can inspire users to tag their photos and Hypertext ’06, ACM Press (2006).
give them guidance for how best to annotate. Based on our 12. Marshall, C. Toward an Ecology of Hypertext Annotation. In
observations, we believe that people are more inclined to Proc. Hypertext ’98, ACM Press (1998).
tag their content when they are given the right incentives 13. Naaman, M., Paepcke, A., and Garcia-Molina, H. From Where to
and affordances for annotation. What: Metadata Sharing for Digital Photographs with Geographic
Coordinates. In Proc. CoopIS ’03, ACM Press (2003).
FUTURE WORK 14. Naaman, M., Yeh, R.B., Garcia-Molina, H., and Paepcke, A.
The taxonomy of tagging motivations would benefit from Leveraging Context to Resolve Identity in Photo Albums. In
additional data from more ZoneTag users over a longer Proc. JCDL ’05 (June 2005).
period of time, especially as ZoneTag is incorporated into 15. Okabe, D., and Ito, M. Everyday Contexts of Camera Phone Use:
more users’ daily practice. In particular, we would like to Steps Toward Technosocial Ethnographic Frameworks. In Mobile
follow up on emergent practices among groups of friends Communication in Everyday Life: An Ethnographic View,
forthcoming.
who all use or are familiar with ZoneTag, including real-
time collaborative tagging and the incorporation of reviews, 16. Rodden, K. How do People Organise their Photographs? In BCS-
IRSG Colloquium on Information Retrieval, 1999.
inside jokes, and ad-hoc photo collection coordination into
17. Rodden, K. and Wood, K. How Do People Manage Their Digital
tags. We would also like to test the taxonomy on additional
Photographs? In Proc. CHI ’03, ACM Press (2003).
Flickr tagging data as well as on other tagging systems.
18. Salovaara, A., Jacucci, G., Oulasvirta, A., Saari, T., Kanerva, P.,
ACKNOWLEDGMENTS Kurvinen, E., and Tiitta, S. Collective Creation and Sense-Making
of Mobile Media. In Proc. CHI ’06, ACM Press (2006).
First and foremost, we would like to thank our participants
for their cooperation and patience. We would also like to 19. Sarvas, R., Herrarte, E., Wilhelm, A., and Davis, M. Metadata
creation system for mobile images. In Proc. MobiSys ’04, ACM
thank to Shane Ahern and the rest of the ZoneTag team at
Press (2004).
Yahoo! Research Berkeley for their assistance. Vlad
20. Sen, S., Lam, S.K., Rashid, A.M., Cosley, D., Frankowski, D.,
Kaplun developed the Flickr photo-elicitation tool, based on
Osterhouse, J., Harper, F.M., Riedl. J. tagging, communities,
the tool developed in [23]. Finally, Mirjana Spasojevic, vocabulary, evolution. In Proc. CSCW ’05, ACM Press (2005).
David A. Gibson, and Ka-Ping Yee provided invaluable 21. Shneiderman, B. and Kang, H. Direct Annotation: A Drag-and-
comments on early drafts of this paper. Drop Strategy for Labeling Photos. In Proc. INFOVIZ ’00, May
2000.
REFERENCES
1. Flickr, [Link] 22. Van House, N., Davis, M., Ames, M., Finn, M., and Viswanathan,
V. The Uses of Personal Networked Digital Imaging: An
2. Frohlich, D., Kuchinsky, A., Pering, C., Don, A., and Ariss, S. Empirical Study of Cameraphone Photos and Sharing. In Ext.
Requirements for Photoware. In Proc. CSCW ’02, ACM Press Abstracts CHI ’05, ACM Press (2005).
(2002).
23. Van House, N. Interview Viz: Visualization-Assisted Photo
3. Girgensohn, A., Adcock, J., Cooper, M., Foote, J. and Wilcox, L. Elicitation. In Ext. Abstracts CHI ’06, ACM Press (2006).
Simplifying the Management of Large Photo Collections. In Proc.
INTERACT ’03. IOS Press (2003). 24. von Ahn, L. and Dabbish, L. 2004. Labeling Images with a
Computer Game. In Proc. CHI ’04, ACM Press (2004).
4. Golder, S., and Huberman, B.A. The Structure of Collaborative
Tagging Systems. Tech. Report, HP Labs, 2005. 25. Wenyin, L., Dumais, S., Sun, Y., Zhang, H., Czerwinski, M., and
Field, B. Semi-automatic image annotation. In Proc. INTERACT
’01, IOS Press (July 2001).