Submissions:2015/Reimagining the article submission process

Title

Outreach

Workshop

Author

wikipedia2015@idea.org

idea

IDEA.org

Abstract

The current process of creating articles suffers from historical, procedural and usability problems.

Notability: The way that people define and imagine worthiness of a topic, and the kinds of primary sources which they think would would justify it, differ from the secondary and tertiary sources which are the bread and butter of Wikipedia notability and citations. In an encyclopedia with nearly 5 million articles in English, and most traditional encyclopedic topics already covered, most new article proposals are about people, companies, and less-notable topics.

Process: The process of using Wiki syntax is confusing and should not be the barrier to entry for introducing short, legitimate topics. The typical formula for various article types is elusive. The process of creating an AfC, or making a new page that could quickly enter deletion proceedings is mired in obtuse and user interface, overwhelming bureaucracy, excessive, often-contradictory regulations, and inconsistent application of rules depending on the whims and biases of the editors reviewing AfC or pursuing deletion, and a feedback style that would seem harsh and arrogant to most outsiders. It's much easier for an editor with AFCH to click a few buttons to decline a new article (~10 seconds) than it is for a new user to laboriously attempt to create an articles (hours). Review and re-review tends to involve different editors, confusing new authors with mixed messages.

COI vs. NPOV: Editors are trigger-happy to throw indefensible COI claims, so editors go underground, and a disproportionate number of new articles are submitted by single-use accounts which attempt to cloak the association of the author. Moreover, the bureaucracy is to dense that many resort to hiring marketing firms, and a cottage industry of sycophantic marketing freelancers who ghostwrite or post. The reality is that for most new topics, which are not nobel prize winners, fortune 500 companies, or incredible famous topics, the only (and often the the best champion) to create those new articles is affiliated with the topic. Instead useful feedback could lessen the need for hiring consultants, and focusing on NPOV would be more productive than focus on COI.

Copyrights: It is difficult and confusing for authors to know how to turn over their text content or images for CC Attribution-ShareAlike use. A good example might be repurposing part of a scholarly article (which the author owns) or thesis.

This workshop would try to re-envision a new-article process as more of a multistep process, rather than cycles of submission/rejection. The interface would better matches the expectations, needs and circumstance of new authors, encourage transparency and honesty, acknowledge the de-centralized nature of AfC/AfD review, and the fact that most articles are rejected the first time. The process could be more iterative by design, starting with a notability claim/test (e.g., a 1 paragraph application which attempts to establish notability with 1-5 key citations), clear and simple tutorials which meet 90% of the common mistakes, and 90% of needs, and an iterative process that reviews the idea and coaches the author over a period of days or 1-2 weeks (which would actually be faster in the end), also triaging inappropriate topics a lot earlier in the process. The goal would be to get good-enough, short Class-C articles into the system, with less frustration and disillusionment from new authors, and less burden on editors. While at the same time, discouraging inappropriate submissions earlier in the cycle.

There is no long-term implementation plan at this time. The output would mostly likely be a series of wireframes, workflow descriptions, or key points for a tutorial.

Length of presentation

75 min.

Special schedule requests

Will you attend WikiConference USA if your submission is not accepted?

Maybe.

Interested attendees

If you are interested in attending this session, please sign with your username below. This will help reviewers to decide which sessions are of high interest. Sign with four tildes. (~~~~).

1. Tokyogirl79 (talk) 04:58, 20 August 2015 (EDT)
2. Add your username here.