Rarely our exper­i­ence with a news story is a dir­ect one, therefore our point of view on a story is driven by habit and con­ven­tions and not created through a first hand exper­i­ence. To under­stand a such story you have not to only under­stand the elements that com­poses it, but the rela­tion­ship between them. Under­stand­ing means decon­struc­tion and reconstruction.

Design

PLESPER works at two levels: the usual macro level is concerned with an article (or any web addressable item) as unit (regarded as a coherent point of view), the micro level dissects an article into smaller segments (ideally at the level of facts).

The macro level navigation is what we’re currently used with as passive readers. The micro level is needed for active reading, for ques­tion­ing assump­tions, con­sid­er­ing altern­at­ives, ques­tion­ing the trust­wor­thi­ness of the authors and their sources. An active reader will recon­struct his crit­ical point of view for deeper understanding.

PLESPER is organised around the idea of shareable space, akin to Etherpad, as data and not as live interactions, each person even when collaborates on the same space may have different views, but the same items.

At macro level PLESPER allows the people sharing a space to add items (by URL or a web search), view a thumbnail representation of them, filter and sort them by various dimensions (location, names, social tags).

A zoomable user interface (ZUI) allows switching from macro to micro level while keeping the navigation coherent.

At micro level each paragraph or possible meaningful sentence can be individually addressed, selected and related with ones from other articles by various means:

• simple similarity (from n-​grams to named entities)
• sentiment analysis
• similar facts (that answer the same question)

Additionally a selected segment can be promoted (quoted) to the macro level in order to be used in constructing relations.

A quote (for example a tweeted quote with a link) can be related not only to its original context if provided but for brevity sake to the minimum context needed to decrease its ambiguity.

Once the initial items were sorted, filtered and dissected; they can be connected in a graph to convey sequence, relations, etc.

Spaces can be shared fully or read-​only simultaneously (different capability URIs), they can be imported (cloned or synced); allowing the construction of various interaction processes.

You can test the prototype as shown in the video at PLESPER​.com (server temporarily offline)

Technical notes

• Item representation is based on oEmbed protocol when available otherwise a Readability–like scraper is employed to extract the meaningful part of a referred page
• Location, names, etc. are extracted via an external service like Thomson Reuters’ OpenCalais
• Comments & annotations should be based on existing (semantic) web technologies and activity streams
• The need for identity & access rights management can be delegated to 3^rd parties through the use of capability URIs.
• Source code http://​github​.com/​L​a​u​r​i​a​n​/​P​L​E​S​PER

Brief

PLESPER is a platform for collaborative storyboarding that does not impose a specific process. People can create, re-​use (fork), mix elements from various storyboards (spaces), share and comment on them by using their exiting (and future) social networks. Several envisaged uses are:

• collaborative drafting of stories with full or partial public collaboration
• published along an article to support it and gather feedback (comments, storyboard re-​use, etc.)
• evolve articles based on the public interaction with the storyboard (at the discretion of the editor)
• create meaningful articles’ connections with other ones (as they may connect at a more abstract/​storyboard level)
• back data visualisations with more in-​depth discussions and datasets (allowing the community to create alternative visualisations and import additional datasets)

One of the most important feature is that there are no specific requirements for the exist­ing sites other than be accessible over the web, PLESPER will work over old pages on the web, it will work with legacy and future CMS sys­tems, it will work over Twit­ter, Face­book, G+ and whatever will be inven­ted; because it is based on what the web is made of: links.

Relevant blog posts:

1. The Perplex & Other Stories (the initial idea)
2. News Stories & Interaction (where I initially explore some existing news spaces navigation metaphors then I introduce PLESPER)
3. Collaboration in PLESPER (a simple playlist analogy)