What aren’t we going to build?

maxcutler: 3 journo devs and 6 hours to work. Please give us project ideas! Tomorrow with @danielbachhuber and @davidestes

The question isn’t what are we going to build, but really what aren’t we going to build?

Open Assignment Desk

The Open Assignment Desk (formerly known as the Virtual Assignment Desk) is a tool for leveraging openness in the story creation process. Hat tip to Jay Rosen and Dave Winer for talking about the left side of the same idea in episode #12 and episode #18 of Rebooting the News.

It brings the funk in stages.

The first is to fulfill the needs of the newsroom in regards to managing story workflow. Stories can start as pitches, get approved and become drafts, and, once completed, go through the editing process to become published pieces. The Open Assignment Desk fills things in by tracking all of the meta data associated with this process, including when the story is due, whether there will be associated photography, the location of the story, etc. Each newsroom would be able to fine-tune their workflow as well. If the publication was going for speed, then their editing process might just be one stage. If the newsroom was more concerned about the accuracy of their content, then they might have a three stage editing process. This idea isn’t really that unique; Max has a product he’s working on for the Courant News CMS called Nando, and CoPress has a currently stalled project called Edit Flow that will soon be back on its feet.

Once the newsroom has adopted the tool, then you roll out stage two: the ability to make this editorial flow public. The newsroom has granular control over which parts of the editorial process are transparent. If they decide to be open about the decisions within the newsroom, the news organization can build engagement with their community and, ideally, the community develops a greater sense of ownership of its journalism as process.

The third act is to allow the community to contribute to the reporting flow in a meaningful way. There are two specific things I’m considering at this point:

  1. The editorial process is exposed in such a fashion that, at any point along the way, the community can contribute content they think is relevant and useful. “At that concert last night?” the news organization asks. “We’re working on a story about it and would love to feature the best of the photos.” The community can also submit notes and opinions on the story in progress much like commenting on the final product. The comments they add when the story is in progress can be used (and linked to) as sources for the story.
  2. The community is able to pitch story assignments. If they see something they don’t think is being covered, they can pitch it as a story and start on the reporting process by identifying the questions that need to be asked, attaching images, and so on. One positive side effect of this functionality would be the ability to dump “all” of the stories that the community thinks need to be covered and then deduce the percentage that the news organization is actually covering. The abstraction of this is the ability to identify all of the information the community thinks it needs, and then use that as the foundation for the reporting process.

News Community Relationship Management (nCRM)

News CRM acts as an open, participatory rolodex. The goal is to capture common knowledge about sources based both on what the journalists of a news organization know, as data generated by how the source interacts with the news organization.

The best way to describe this is by doing a mental fly-through. When a journalist is interviewing someone the system has never seen before, they create a new profile in the system. The profile contains fields for contact information, location, occupation, etc. and a free form wiki text area for notes about the person. The journalist doing the interview can also create an association between their text notes, audio file from the interview, and other content within the CMS to the source profile. By doing this, journalists who need to pull information about the source in the future will have access to the sum of news organization’s knowledge.

Assuming the source interacted digitally with the news organization on a regular basis, the CRM application would also be responsible for extending their profile by tracking those interactions. For instance, in step one the journalist would record any or all email addresses regularly used by the source. If the source comments on an article where they are quoted, then the comment could be flagged as a clarification in relationship to their profile. If the source comments on another article, then the tags of that article would be applied to the source profile as topics of interest. Based on how the community reacts to the comment, by voting it either up or down, the source’s authority on the topic would change dynamically.

The goal with this tool is really to structure information about a news organization’s community such that when the reporter needs to do interviews on, say, water, they have an entire database of the “right” people to seek out.


Greg Linch August 7, 2009 Reply

This should be livestreamed from each of your webcams. On uStream — so I can watch/listen on my iPhone.

It should be Epic.

Joey Baker August 7, 2009 Reply

Yea… okay, I might have wet my pants a little bit at the nCRM concept. Privacy is sorta a major (unresolved) issue, but who cares? Build the dynamic comment-expert-tracker, with the … oooo! Business model!

Build this software with the intention of indexing all commenters. Provide the data free to everyone, sorta a backtype meets wikipedia. Then freemium-ize it! If reporters want to add in interviewees, that’s fine. If you want to have more than 1 reporter, that’s $xxx/mo. If you want to keep details you enter private, that’s $xxx/mo.

/brain dump

Dudes! I’m stoked to see what ya’ll come up with!

Andrew August 7, 2009 Reply

Sweet! Like Joey I can’t wait to see what you guys come up with. The CRM idea particularly interests me because it’s just something that I don’t really see as existing out there.

Will you guys just be creating specs tomorrow or will you actually start writing the code for these projects as well?

Daniel August 7, 2009 Reply

To be honest, I don’t have a clue what we’ll actually end up doing tomorrow. Max tweeted about doing a code sprint a few days ago, I asked him whether he wanted to come down to Portland, and it sounds like David was interested too. If the project is small enough, we might do some building; otherwise, it might just be sketching ideas out further.

Bill Fitzgerald August 8, 2009 Reply

Hello, Daniel,

The Open Assignment Desk could be built in Drupal using existing tools, possibly without writing a line of new code.

The Workflow module could support different workflow states for articles, and could also control access to data based on workflow state. The metadata (supporting photography, due dates, etc) could be set up as a string of connections that could be browsed, and the community involvement piece could allow site members to contribute to the piece by adding content as the editors wanted/needed.

You could really add some special sauce and mix in some Solr search goodness, and you would have faceted browsing through this material, as well as content recommendation (find similar articles, which would allow for connections to be made between assets that weren’t considered related).

You could even take this a step further and include geotagging into the mix, and allow the news to be represented geographically.

And if you looked at importing rss feeds of local events, and/or of other sources of local news, you would have a great starting point to allow the most interesting stories and events to percolate upwards.

And to be clear: all of the functionality I describe in this comment can be built on top of stable, existing code from the Drupal community.



Leave a Reply