Course:LIBR559M/When Push Comes to Not-Shove, but Pull

From UBC Wiki
Push/Pull Technology
Wiki.png
LIBR 559M - Social media for information professionals
Section: Section 98B
Instructor: Dean Giustini
Email: dean.giustini@ubc.ca
Office:
Office Hours:
Class Schedule:
Classroom:
Important Course Pages
Syllabus
Lecture Notes
Assignments
Course Discussion
[[Category:]]

Contributors

  • Campbell, Tristan
  • Ferguson, Greg
  • Graboski, Elizabeth
  • Stathers, Kim

Introduction

The purpose of this wiki page is to give a general overview of push and pull technology as well as the technological uses and benefits. The popular push and pull platforms used in libraries and archives are discussed citing how each impacts client service delivery and librarian professional development.

Definitions

Push technology describes a form of internet communication that automates the delivery of information from a server to a client. The request for a given transaction is initiated by the publisher and is delivered by "pushing" it to the client in real time in a continual running data stream instead of waiting for the client to specifically request it.

Pull technology is the opposite of push technology. The request for an information transaction is initiated by the client and is delivered "pulled" from the server to the client one item at a time.

Technology

Push Technologies

The content that is delivered is selected differently for push and pull technologies. In a push scenario, the server "publishes" content into classes, and the client "subscribes" to classes that it wants to have streamed to it. Content is not delivered to individual clients, but rather to the population that subscribes to a given class. So, the publishers and subscribers are decoupled because there is not a 1:1 correlation. This decoupling allows for greater scalability and a more dynamic experience for the client. Transactions are immediate and efficient, but the process can use up a significant amount of bandwidth. Comet programming (AKA: HTTP server push or HTTP streaming), uses a persistent connection between the client and the server to push data. Variations of this process of extended server-client connections are the basis for both pushlet and XMPP via HTTP. In pushlet technology the maintained connection to the server keeps the browser loading continuously and the server uses javascript to push new data to the client. XMPP is a common protocol used for instant messaging, such as Google Talk and Live Journal's Instant Messaging feature. XMPP via HTTP, which is often useful when the client is behind a firewall which restricts XMPP, uses HTTP binding to maintain connection to the server, and uses the GET and POST functions of HTTP to send and receive messages.

Pull Technology

In a pull scenario, the client (device) asks the server at select intervals if there's any new content, and if so, the server delivers it to the client. This is called a delayed "store and forward" methodology. There are two ways in which pull technology can simulate push technology; frequent pulling and long polling. Frequent pulling involves sending requests to the server frequently enough that the user experiences the pulls as though the data were being pushed, such as a POP3 email client pulling from the server every minute or two. The frequency simulates push technology but is very inefficient. In long polling the server holds the request made by a client until their is data to be sent, and each time data is sent the client sends a new request, so the server always has a request to answer when it receives data headed for the client. This is not a true push but the effect is the same, and it is much more efficient than the repeated requests of frequent pulling.

Uses of Push / Pull Technology

Pull technology forms the basis of computer networks, and thus the Internet. Web browsers use HTTP requests, a pull technology, to request website content from web servers. Traditional e-mail clients, such as Microsoft Outlook, rely on POP3 or IMAP pull requests to get mail from mail servers. Web feeds (such as Atom/RSS) are a more recent development which also rely on clients pulling data from central server. The problem with pull technology is bandwidth inefficiency; clients must continually poll servers to find out if there is new content. A large amount of clients that frequently poll a server for new content will create an enormous amount of data traffic without any actual useful content exchange.

Push technology is as equally common and important as pull technology. Text, voice and video messaging (such as Skype) all use push technology; the server receives the information from one client and immediately pushes it out to the other client(s). E-mail also uses push; the e-mail is pushed from the sender to the server, and from server to server. Only the final transmission, from server to recipient, relies on pull. Even this is changing; Blackberry, iPhone, Andriod phones and other smartphones support push e-mail from various providers. The web itself is not completely pull; HTTP can be exploited in various ways to open a push connection, allowing servers to send information to web browsers without explicitly being asked for it.

Applications For Libraries and Archives

Push Technologies used by Libraries and Archives

One of the first forays into push technologies hosted by libraries was in the development of consortia for electronic scholarly publication managed through electronic commons. This allowed libraries to become Internet providers of this information. One example of this is the Oklahoma State University Electronic Publishing Center (http://digital.library.okstate.edu/about.html).

Similarly, libraries and archives in a corporate or academic setting can use push technology to send out updates of documents within an institutions closed Intranet.

Another service that libraries and archives provide to patrons via push technology is Instant Messaging (IM), which allows direct contact and transfer of information from a reference librarian or archivist to the user.

Libraries are more commonly the client of push technology as the recipient of requested information in the form of digitally delivered subscriptions of databases and ebooks.

Pull Technologies used by Libraries and Archives

  • Facebook, Twitter and RSS feeds are the most popular pull technology platforms used by libraries and archives.

Information Service Delivery

Twitter seems to have the most promise as an information broadcast tool combining marketing and current service alerts. Using the "Hub and Spoke" model of organization, Twitter easily replicates and enhances the traditional delivery of information service by offering an unlimited level of complexity and granularity in the accounts that programs, services and individuals might create. With the addition of the Tweethopper, a tool that controls multiple accounts, automates the conversion of an RSS feed into tweets and will auto-follow followers, management of an organization’s Twitter accounts become time efficient.

Professional Development

The pull technology that is preferred by librarians/archivists in professional development is Twitter and is achieved primarily by following feeds from professional organizations and high profile library and information professionals active on Twitter. The interaction with professional peers on a global level contributes to improved services and programs through a learning process of strategic “follow” relationships.

References