Posts Tagged ‘ CMS ’

Parallactic Drift beta is now live!

About time! I’ve been working on this project and finally, I have something posted online to show for it.

Right now Parallactic Drift stands more as a proof of concept of a scalable AJAX framework that can fit into any CMS that can automate its compiling XSLT scripts (through either Cron or Cron-like functions) and a better search function.

The framework is entirely client-side, using jQuery to hide and present search query-matching stories in individual modal windows that float above a fixed map with location points. As a result, it’s a pretty big page with a slow search function and other limitation (i.e. the page does not contain my favored fade others function for emphasis because of the sheer number of news items). The news items come from a few weeks of gathering all of the RSS feeds from the,, and So individually, the news items number in the thousands. Continue reading


API development sheds light on a new workflow

Making a stand-alone scraper and XML readers seems to be the vogue nowadays. I met news API developer for The New York Times Derek Willis last week, and he brought up the good point that many online developers working in journalism had to use APIs and other querying services for their own publications because they are usually not given admin access to the publication’s electronic database.

Like Derek, I want to help journalists solve redundancies within their information gathering and distribution models. All of the hard data that many of my hard-working colleagues gather (i.e. names, ages, dates, etc.) should be stored at the most granular level within a relational or hierarchical database should be easily reused and accessible to everyone, not just journalists. Practical sorting of individual articles, i.e. by relevance, relies on an article’s meta-data, the most accurate of which is derived from the article’s most minor elements. Continue reading

Operational difficulties of querying semantic information

Folder Structure and Workflow for Parallactic DriftIn order to implement an efficient system of organizing news items, content providers must label information in a common way within each platform, be it RSS, blogs or web sites. Standards in fact do exist for XML tagging for news sites. Several web consortia exist (including W3C and NewsML) to ensure that a single format is followed, and that information flows freely between publications and reaches more users.

Perhaps it’s because this principle of “free-flowing information” seems in itself counter-intuitive to how traditional publications share news items, but an inconsistent style stifles any RDF standard across different publications. Even if designs remain idiosyncratic, as they should, the semantic tagging of information, in HTML and XML should not deviate too much from an agreed-upon standard. Continue reading

Semantic relevance and news web archives

Last semester I contributed to a report for my Applied Research in Content Management class wherein each student identified a problem and a solution with, a student-run news site. For my section in the report The NewsHouse Optimization (Fall 2009), I focused on the importance of interlinking news items by relevance rather than chronology, offing it as a solution for drawing readers further into the site’s content.

TheNewsHouse, which runs on a Drupal CMS, uses many tools to interlink stories such as related tags and bylines. When clicked on, however, the site ranks stories that match these tags or keywords by date published rather than strict relevance. Individual stories also do not contain links to news items with similar information but instead to similar bylines, and non-bouncing visitors tend to concentrate within these byline-linked stories according to TheNewsHouse’s analytics.

To ensure the maximum potential viewing of their content, news sites such as must conform to what users have come to expect. As I mentioned in my previous post, Google and other powerful search engines now retrieve news and information that is more and more relevant to the user’s particular interest. To cater to these particular interests, the news site’s management must optimize the linking of stories by associating news items to other news items that are as semantically relevant as possible. Continue reading