A STRAIGHTFORWARD APPROACH FOR ONLINE

ANNOTATIONS: SPREADCRUMBS

Enhancing and Simplifying Online Collaboration

Ricardo Kawase and Wolfgang Nejdl

L3S Research Center, Leibniz Universität Hannover, Appelstr. 4, 30167 Hannover, Germany

Keywords: Annotation, Social Media, Social Network, Online Collaboration, User Interface, SpreadCrumbs.

Abstract: Countless user studies and everyday observations have shown that individuals make annotations while

reading - highlighting, circling and underlining important parts of the text, moreover adding written

comments. Since the Web became the biggest accessible source of information, many of the reading

activities happens online in the browser. In this sense, it is expected that the individuals would keep their

annotation behaviors, provided that the appropriate tools are available. Although several Web annotation

projects currently exist, it is difficult to identify the most prominent in the field. With SpreadCrumbs, we

simplify the annotations actions and the social navigation support. SpreadCrumbs users can add in-context

annotations to any webpage with minimum cognition load, as they would do when reading a paper; in

addition SpreadCrumbs enhances online collaboration and provides mechanisms to support social

navigation by means of existing social networks. It allows the users to freely express themselves and to add

any desirable substance to the resources. Technically, annotations carry valuable information about the

content, more than bookmarks or tags, having a greater impact on collaboration and search for re-finding.

SpreadCrumbs exploits all these advantages with an intuitive and easy-to-use user interface.

1 INTRODUCTION

We understand annotation as some extra

information attached to a resource that can assume

many different forms. In-context annotations may

not only help you later but may be useful as well for

other future readers. Eventually, scribbling is

extremely common during reading activities. In

some user driven tests O’Hara and Sellen (O’Hara,

1997) demonstrated that most of the subjects used

annotations to help understand the text and to aid in

the future task of writing. In an impressive field

study on annotations in college textbooks, Marshall

(Marshall, 1997), (Marshall, 1998) managed to

identify patterns in annotations, statistics and further

more describing and classifying the many forms of

annotations such as: signalling for future attention,

memory aiding, problem-working, interpretation,

progress tracing in narrative and so on.

Given that the Internet is the largest source of

information nowadays, it is expected that a lot of the

readings occur online; consequently Web annotation

would be an expected feature on the Internet.

However, no annotation system so far has shown

nimbleness, perspective or has survived the first

years of existence, nevertheless, it has been far

discussed the importance of annotations for

comprehension and also the benefits for

reading/writing proposes (O’Hara, 1997). Given the

absence of any dominant mature annotation system,

it appears that there is still no generally accepted,

concrete method for straightforward online

annotation. In order to understand such problem –

and the user’s preferences for tagging and

bookmarking systems over annotation systems -

we’ve developed a simple, easy to use and

straightforward system that supports Web annotation

with basic features of annotation, bookmarking and

social navigation support. In this paper we introduce

this promising system explaining its usage and

functionalities.

The rest of the paper is structured as follows: In

Section 2 we go through some related work. In

Section 3 we describe the SpreadCrumbs system

closing with the future work and conclusion.

407

Kawase R. and Nejdl W.

A STRAIGHTFORWARD APPROACH FOR ONLINE ANNOTATIONS: SPREADCRUMBS - Enhancing and Simplifying Online Collaboration.

DOI: 10.5220/0001824804070410

In Proceedings of the Fifth International Conference on Web Information Systems and Technologies (WEBIST 2009), page

ISBN: 978-989-8111-81-4

2 RELATED WORK

The first group of related works is the existing and

past commercial tools for web annotation.

ThirdVoice

was probably the first expressive

commercial Web annotation tool. It was a plug-in

for Internet Explorer 4 and Netscape Web browsers

which allowed the users to publicly annotate any

webpage. The “in margin” written annotations were

visible to any user of the application that accesses

the site. It is not completely transparent where

ThirdVoice failed but the service was discontinued

on April 2001. Some other old discontinued

commercial systems Hypernix, NovaWiz, utok,

Zadu followed the same steps with less public

attention. More up-to-date systems Fleck

SharedCopy

, Diigo

bring a new air for the

annotations scenario. They grant tagging, re-finding,

collaboration, social navigation and annotation itself

working in the same way as the old tools - a plug-in

for the browsers. Still, none of them reached a

considerable impact level in the Web as it happened

in social networks, folksonomies and tagging

systems.

In addition to the commercial tools, several

research projects aim to enhance Web collaboration

by providing annotation capabilities. We have been

through some of these works to try to understand the

evolution in the Web annotation scenario.

The Anchored Conversations system (Churchill,

2000a) provides a synchronous text chat window

that can be anchored to a specific point within a

document. It is presented as a post-it note and can

also be used for re-finding by the system search

option. In this case the collaboration occurs during a

synchronous chat. Like the Anchored Conversations

we understand that the most appropriate metaphor

for transient annotations is the post-it notes.

Fluid Annotations (Zellweger, 2001) supports in-

context annotations and it is an extension of the open

hypermedia Arakne Environment (Bouvin, 1999).

But different from other researches, the studies and

evaluations are mostly presentation of the

annotations, as seen in (Zellweger, 2000) and

(Zellweger, 2001) in terms of visual cues,

interactions and animated transactions. Their

evaluations give valuable material for annotations

manipulation and usability, however, their approach

of “between lines” annotations disrupts the original

http://www.ThirdVoice.com (March, 2000)

http://www.fleck.com

http://sharedcopy.com

http://www.diigo.com

layout of the annotated content besides the

distractive animation transactions.

In the end, all attempts, projects and commercial

tools aim to enhance communication and

collaboration among the users independently of the

task. Putting together all those systems there is a

common understanding of the potential value

provided by annotations nevertheless few has been

used in large scale to gather enough data to

understand the user’s behaviors and existing patterns

during online task-free annotation practice.

3 SPREADCRUMBS

SpreadCrumbs is an in-context Web annotation

system. The basic idea is that using SpreadCrumbs

the users can annotate Web resources with keywords

or sentences and share these annotations over a

social network. We can compare SpreadCrumbs to

social bookmark systems such as del.icio.us.

GroupMe!

with the slightest but crucial difference

that the annotations are displayed on a layer over the

Web resource (in the context). We use a very

common visual metaphor for annotations, reminders

and alerts: post-it like notes. The annotations (which

we will identify as a crumb) are added with

minimum effort. The right click context menu shows

an option “Add Crumb”. Instead of creating our own

social network we decided that the right strategy

would be to bring one’s service to the users and not

the other way around. Due to that we chose to build

our social infrastructure on Facebook

platform. In

our quest for answer the absence of annotation in the

Web, this strategy covers what maybe the first

dilemma – many users are not very confident

subscribing to new Web sites services or installing

new applications.

SpreadCrumbs consists in two different

implementation packages, the Facebook application

with the SpreadCrumbs server pages and the

browser (FireFox) plug-in.

3.1 SpreadCrumbs Characteristics

On SpreadCrumbs we have two different concepts:

topics and crumbs where each topic contains at least

one crumb. The post-it note metaphor is a topic with

n crumbs. Each topic on SpreadCrumbs is associated

to an URL and it can assume different purposes. A

http://delicious.com

http://groupme.org

http://www.facebook.com

WEBIST 2009 - 5th International Conference on Web Information Systems and Technologies

408

private topic where the user post it to herself is an

annotation used as a bookmark, a reminder or some

extra information about the resource. A shared topic

between many users is a shared bookmark and also a

reminder for the people involved in the topic.

Moreover it assumes the behavior of a discussion

group or a forum on the defined context. In this last

case the users involved can add new crumbs to the

topic.

The post-it note contains the topic creator, the

participants of the topic, the topic itself and the

crumbs texts. The mouse-over tooltip shows the

dates of each crumb. From the post-it interface the

user is able to delete her crumb, reply the topic,

change her status in the topic and follow her trail of

crumbs or the trail left from other users as shown in

figure 1. The users names involved are links to the

user profile on Facebook and the mouse-over tooltip

shows the user status in the topic - additionally,

disconnected users have their names strikethrough.

The SpreadCrumbs

application on Facebook is

the user’s “mailbox”. We have different pages for

the different status of the user in a topic. On each

page the crumbs and topics are grouped by the

website URL. There the user can read and delete her

crumbs, reply topics, change her status in a topic and

check the statistics of the crumbs. Finally, an

internal search engine that searches all

SpreadCrumbs’ content lets the users re-find the

annotated pages.

Figure 1: A crumb on WEBIST webpage sent from

Ricardo to Wolfgang.

Social navigation support (SNS) describes

techniques to guide users through specific chosen

http://apps.facebook.com/spreadcrumbs

resources (Brusilovsky, 2001). Past systems relied

basically on collection annotated by experts. The

AnnotatEd (Farzan, 2006) system authors introduce

two types of SNS: traffic-based SNS and annotation-

based SNS. Our model fits on with the annotation-

based style, wherein every annotated page becomes

a step in a trail.

In the current version of SpreadCrumbs each

crumb contains the name of the author together with

a “back arrow” and a “forward arrow”. These arrows

are links to the previous/next crumb (in temporal

sequence) that the author wrote for the user being

considered. The current model allows the users to

create trails with content and ordinality relevance.

SpreadCrumbs interface provides a much more

concrete model for browsing with SNS. The user

follows exactly the trail given to her, different from

other systems where the SNS is just a collection of

links annotated or bookmarked from another user.

Still, our social navigation support model is not

totally defined. One problem identified is the

difficult for distinguishing different trails: when a

trail ends and when a new one begins.

4 FUTURE WORK AND

CONCLUSIONS

So far we have not evaluated the SpreadCrumbs

with respect to a particular annotation task or user

groups. Our primary observations show that the tool

has been favourably received by the users, although

further analyses are required to fully evaluate with

respect of usability, performance, and user

satisfaction. We suspect that SpreadCrumbs may be

useful for bookmarking activities and will show

better results in the tasks of re-finding bookmarks

and locate relevant content within the bookmarked

pages, being valuable for everyday online

annotations, e-learning purposes and general

collaboration.

We are currently running the first part of user

studies tackling online collaboration through

annotations and information retrieval from

previously annotated pages. We still miss a second

round of the studies with the same participants to

collect the complete results. We have strong beliefs

that this evaluation will demonstrate the benefits of

annotations over tags and bookmarks. Further

evaluations are constrained to the number of user of

the tool and also the frequency of usage. We hope

that our simple mechanisms and intuitive interfaces

will motivate the users to annotate and collaborate.

A STRAIGHTFORWARD APPROACH FOR ONLINE ANNOTATIONS: SPREADCRUMBS - Enhancing and

Simplifying Online Collaboration

409

All the results regarding the evaluation will be

posted as soon as they are ready.

Regarding implementation we want to make

SpreadCrumbs available for all, which means to

have different alternatives in some of our current

implementations constraints. At the moment

SpreadCrumbs only supports Facebook social

network. This can be extended to incorporate some

others social networks such as Linkedin

, MySpace,

etc. It is even possible to have an infrastructure

completely disconnected from social networks

where the users can do self annotations and

collaborate with her email contacts. We are still

planning and evaluating the possible benefits and

disadvantage of those approaches.

Some of our early users are researches involved

in some remarkable e-learning and online

collaboration projects. Our collaboration is

producing some great perspectives for connecting

SpreadCrumbs with those works. Our ideas with

GroupMe! (Abel, 2008) researchers are to integrate

both tools giving the users the possibility to gather

online resources through SpreadCrumbs and at the

same time to have manners to group theirs

annotations.

Within TENCompetence

, the prototype

LearnWeb2.0 (Marenzi, 2008) tackles the

functionalities of aggregation and annotation of Web

2.0 resources furthermore, supporting the creation,

storage and exchange of learning objects and

knowledge resources. In the scope of this project,

the initial proposals are to employ SpreadCrumbs

also as a resource collector and annotation tool

integrated with its underlying social network.

In both of these research projects, annotation is

an essential feature, however, such tool to facilitate

this functionality and the “in-context” representation

of the annotations are not supported yet.

Concluding, we presented in this paper a

promising Web annotation tool that addresses

different matters of past projects in the field,

concerning end-user usability, simplification - the

simpler the better - and rescuing user’s existing

social networks for collaboration, in which the users

have the possibility to annotate, in-context, Web

resources in an easy way enhancing bookmark,

information re-finding and cooperation.

REFERENCES

Abel, F., Henze, N., Krause, D., (2008) A Novel Approach

http://www.linkedin.com/

http://www.tencompetence.org/

To Social Tagging: GroupMe!. WEBIST 2008. Funchal,

Madeira, Portugal. May 2008.

Bouvin, N. O. (1999). Unifying strategies for Web

augmentation.Proceedings of ACM Hypertext’99, p

91-100, 1999.

Brusilovsky, P. (2001) Adaptive hypermedia. User

Modeling and User Adapted Interaction 11 (1/2), 87-

110. Claypool, M., Le, P., Wased, M., and Brown, D.

(2002) Implicit interest indicators. In: Proceedings of

6th International Conference on Intelligent User

Interfaces, pp. 33-40.

Churchill, E., Trevor, J., Bly, S., Nelson, L., and Cubranic,

D. (2000a). Anchored Conversations: Chatting in the

Context of a Document. Proceedings of the 2000

ACM Conference on Human Factors in Computing

Systems (CHI 2000).

Farzan, R., Brusilovsky, P. (2006). AnnotatEd: A Social

Navigation and Annotation Service for Web-based

Educational Resources. In: Proc. of E-Learn 2006,

Honolulu, HI, USA, October 13-17, 2006, AACE

2794--2802.

Marenzi,I., Demidova, E., Nejdl, W., Zerr, S. (2008)

Social Software for Lifelong Competence

Development: Challenges and Infrastructure.

International Journal of Emerging Technologies in

Learning (iJET), 2008.

Marshall, C. (1997). Annotation: From Paper Books to the

Digital Library. Proceedings of the 1997 ACM

International Conference on Digital Libraries (DL 97).

Marshall, C. (1998). Toward an Ecology of Hypertext

Annotation. Proceedings of the Ninth ACM

Conference on Hypertext and Hypermedia (Hypertext

98).

O’Hara, K., Sellen, A. (1997). A Comparison of Reading

Paper and On-Line Documents. Proceedings of the

1997 ACM Conference on Human Factors in

Computing Systems (CHI 97).

Zellweger, P., Bouvin, N. O., Jehøj, H., Mackinlay, J.

(2001). Fluid annotations in an open world.

Proceedings of ACM Hypertext 2001, pp. 9–18, 2001.

Zellweger, P., Regli, S., Mackinlay, J., Chang, B. (2000).

The impact of fluid documents on reading and

browsing: An observational study. Proceedings of CHI

2000, 2000.

WEBIST 2009 - 5th International Conference on Web Information Systems and Technologies

410