Developer Info : Overview

Introduction

Collections combine large groups of similar items on the internet, so we can begin viewing the Web as a "web" rather than a series of isolated pages. As a developer, you can create collections of your own. New collections can be created with no code – they’re only data. This page contains an overview of the collection-building process, as well as links to more detailed pages on specific topics. You can also join our technical discussion to interact directly with the Pivot team. This overview contains the following sections:

• Creating a Collection
• Kinds of Collections
• Architecture
• Challenges

Creating a Collection

Collections are composed of two parts:

• XML – The items in the collection are described in XML, in a ".cxml" file. For details, see: Collection XML Schema.
• Images – The images in the collection are described in Deep Zoom format. For details, see: Collection Image Content.

Creating a collection has four distinct steps:

1. Pick your data – First, pick a set of data to turn into a collection and decide how you want to present it. For tips, see Collection Design.
2. Create XML and images – Once you have your data sources, you’ll need to describe it in Collection XML (CXML) and transform your images to the Deep Zoom format. You can use our Excel-based tool, or you can build your own tools. See Collection XML Schema, Collection Image Content, and Collection Design for detailed information.
3. Host it – To share your collection with others, host it on a web server. For more information, see Collection Hosting.
4. Share it – When it’s hosted, post a link in our forum and on your blog!

 Back To Top

Kinds of Collections

There are three primary kinds of collections. They differ primarily in size and their ability to respond to custom user queries. Structurally, they are composed of either previously generated (or static) XML, or XML generated dynamically in response to a query. See the following figure for more information on the three kinds of collections.

Simple Collections	Linked Collections	Dynamic Collections
Difficulty: Easy Size: Up to 3,000 items Examples: Dog Breeds Sports Illustrated Covers	Difficulty: Medium Size: Limited by storage complexity Examples: AMG Movies NFL Teams & Athletes	Difficulty: Hard Size: Unbounded Example: Wikipedia Collection
Details: The most common type of collection. Data is static and loaded all at once. Visuals are static and contained in one DZC (see: Collection Image Content).	Details: Generally used for collections in the several thousands of items. Collection is stored as a composition of inter-linked simple collections. Data is static and loaded one simple collection at a time. Visuals are static and contained in multiple DZCs (see: Collection Image Content).	Details: Used for very large data sets (hundreds of thousands of items or more). Data is dynamic (see Collection Hosting) and loaded in response to a query. Visuals are partially dynamic and contained in dynamic DZCs (see: Collection Image Content).

For more details, see: Collection Hosting.

 Back To Top

Architecture

Conceptually, a collection is just like any other web content. There’s a set of files on a server, and a local client that knows how to display them. In the current web, the files are traditionally HTML and images. In the collection case, the files are CXML and Deep Zoom-formatted (DZC) images. Depending on whether the user browses web pages or collections, the Pivot client will either use the embedded IE rendering engine (Trident) or the collection browser to display the files. See the following figure.

 Back To Top

Challenges

As developers, we hope that you can push the limits of the collection experience and the scenarios it can enable!

We challenge you to:

• Create collections from data in existing online services
• Use collections as a front end to desktop search
• Apply collections as a front end to SharePoint
• Display data sets from biology or natural sciences
• Transform books from the Internet Archive
• Explore information from http://www.data.gov/
• Use Pivot to visualize stock market data
• Apply Pivot to personal photo repositories, or social networks
• Surprise us!

 Back To Top

Feedback

Join our technical discussion to interact directly with the Pivot team. We hope you’ll join this community and share your work.