Abstract - Recently, SNS, blog and microblog have become very popular. A different form of diversified contents, comparing to traditional static HTML web pages, are so easy to be generated dynamically, and can be provided to users by timeline. These contents are generally processed data streams, and have become an important part of the Internet resources. In this study, we try to integrate these dynamically diversified contents, such as twitter stream data, into Ubiquitous Personal Study (UPS), a personalized virtual study proposed in our previous study. In this paper, we introduce and define a set of new metaphors to represent a variety of data streams, such as Drop, Stream, River, and Ocean. We propose a Framework of Organic Streams to meaningfully organize these stream data. We describe the design and implementation issues of a prototype system, and discuss a scenario of using the enhanced UPS to support learning that utilizes twitter data streams. p. 1

In our previous study, we have proposed and designed a supportive information access and sharing environment based on Ubiquitous Personal Study (UPS), a personalized virtual study. We have introduced and defined some metaphors for a set of information collections used in UPS. We have further defined V-Log as a Personal Chronicle, which traces all activities conducted within UPS, and these of Internet accesses via UPS, real-world information behaviors via RFID and UPS. In this study, we try to integrate these dynamically diversified contents, including twitter stream data, into UPS. We define a set of new metaphors to represent a variety of data streams, such as Drop, Stream, River, and Ocean, and propose a Framework of Organic Streams to meaningfully organize these stream data. p. 1

1) ActivityStreams9 ActivityStreams is a draft which creates an activity stream protocol to syndicate activities taken in social web applications and services. ActivityStreams is an extension to the Atom feed format that has been made to express what users are doing on the web. It includes users’ activity with activity object types such as actors, verbs, objects and target elements, which have been applied in Facebook, Google Buzz and MySpace. p. 2

UPS collects individual information together by organizing individual identity and personal information in each UPS. Information that one wants to see or to be shown is brought together in one place. Personal information is organized with metaphors, such as V-Book, V-Bookshelf, V-Desktop and V- Note. V-Bookshelf, V-Desktop and V-Note are also defined as information layers which represent the whole set, the subset that is accessed recently, and the minimal set that is accessed currently. This helps a user to extend the possibility at a maximum to connect the user’s device with a reasonable data size in ubiquitous computing environments. We interconnect a number of UPSs with each other by the XSNS (Cross Social Network Services) module to build a distributed UPS community. UPS environment may also be integrated with functions that aim to establish relationships by exploring friend matching, recommend information, and share information with others. UPS also provides an environment for collaborative activity via the Internet. p. 2

2) ActionStreams ActionStreams is a further extension to ActivityStreams. It not only includes status and activity, but also allows users to take some actions related to an activity. For example, a user posts a friend request that needs an “approve” action. Other users not in this web site who follow this user can click this action to add him/her as a friend instead of logging into that user’s site. Facebook has provided the ability to attach actions links to activities in its newsfeed. p. 2

In addition, a stream has two different trends such as Ripple and Ocean Current. B. Metaphors for Streams z Ripple: Ripple is a collection of the drops related to some topics which are formed naturally in the river or channel. z Ocean Current: Ocean Current is a collection of the drops related to some topics which are formed naturally in the ocean. p. 3

Applying these newly introduced metaphors, chaos of dynamically diversified contents can be turned into meaningfully organized streams, which can be shared and utilized in UPS, as a usual V-Book. Organic Streams can be considered as a collection of drops and streams that converge together in a way they refer or link each other by perceiving the inherent logicality among them. Based on these metaphors, messages posted from users can be seen as a drop. Drops that come from one user to converge together form a stream. The stream of each user can be divided into a series of canals by different tags. Thus the streams of the user and his/her friends form a river. The canals with the same tag from different users form the channel. Finally, all the streams come together to form the ocean. In this paper, we mainly focus on discussing organic streams in the river level. In the river, the drops converging to several different topics form a variety of ripples. Canals can also be formed by a series of drops extracted with ”#” or ”!” tag. In Fig. 2, the drops inside the ripples can be logically and meaningfully organized, which refer or link each other, being followed by the required tags. Through UPS, we can track the related drops and streams with the special tags, keywords, and even users’ profile to explore organic streams, which might include human idea and thinking. In such a way, it can help users to get further understanding of the streams. p. 3

In this study, we introduce and define new metaphors for data streams, which are conceptually shown in Fig. 1, and described as follows. x Drop: Drop is a minimum unit of data streams, such as a message posted to the microblog (e.g., Twitter) by a user, or a status change in SNS (e.g., Facebook). x Stream: Stream is a set of drops in timeline, which contains the messages, activities and actions of a user. x River: River is a set of streams from different users which are formed by following or subscribing his/her followers/friends, and can be extended to followers’ followers. x Ocean: Ocean is the whole set of all the data streams. p. 3

Furthermore, Canal and Channel are defined as simply extracted and organized streams with the messages that have been posted following some special topics. x Canal: Canal is a collection of the drops extracted from a specific user's streams with ”#” or ”!” tag. A user stream can include many canals depending on what extent the user’s interests and topics are diversified. x Channel: Channel is a collection of the canals extracted from different users' streams with ”#” or ”!” tag. Channel is used in the stream level of rivers and oceans. p. 3

Figure 4 shows a scenario of utilizing organic streams for learning support. We assume a class with 200 students, where the professor requires these students to use twitter to publish messages to show if they understand the lecture, to raise questions, and discuss with the professor and/or other students. This will help the professor to monitor and regulate the progress of learning activity. During the lecture, for example, a period of 90 minutes, students may have a variety of questions to raise and topics to discuss with each other. Let’s consider if each student publishes 3 to 5 messages, the total number of messages could be up to 600 to 1000 which is a big number and difficult for the professor to read and analyze in the 90 minutes time slot. Our proposed organic streams in UPS could be expected to help overcome this problem. p. 5

WordPress12 has been used as the core module for our prototype system, which is well known as an AMP (Apache, MySQL and PHP) based solution. WordPress is selected for its easily extensible feature that well fits to UPS prototype construction, since it has been equipped with abundant Plugin to add on new functions, plentiful themes (design) to fit the user’s taste. p. 5

A number of UPS based systems are interconnected with each other by UPS-XSNS module to build a distributed UPS community. Connections among UPS, the popular SNS and microblog are established for providing open API to assess the usability of UPS and also for expanding the UPS community. As a result, users do not think other when posting messages on users’ UPS. Other SNS will get user’s ActivityStreams by following user’s UPS. p. 5

OStatus13 is an open-source draft that builds upon existing open web standards and drafts such as ActivityStreams, PubSubHubbub 14 , Salmon 15 , and Webfinger 16 to enable interoperability among different sites, allowing each site's users to follow and get updated status from the users of the other sites. StatusNet17 is a decentralized microblog, and an open source program that has been implemented under OStatus. In this study, we use StatusNet to follow the remote UPS and create UPS community in the prototype system. We also integrate StatusNet as a plugin of the WordPress. p. 5

Through the UPS, ripples referring to different topics to discover and extract automatically in the channel, which can be analyzed and organized into a meaningful and logical contents by UPS Stream Organizer. They become visible to the professor in real-time, which can help the professor to improve the teaching quality. p. 5

SMOB19 is a platform for open, semantic and distributed microblogging combining Social Web principles and state-of-the-art Semantic Web and Linked Data technologies p. 6

[6] Passant A., et. al. An Overview of SMOB 2: Open, Semantic and Distributed Micro-blogging. In Proceedings of AAAI/ICWSM 2010. p. 6