The Wikipedia Model

What we need that Wikipedia doesn't

• Authority - some way to be able to say "this is right"
• Access control to some content
• Better handling of semantic data (wikipedia needs that too, actually - but we're starting from zero now instead of 5 years ago, and can get it right)
• A stronger editorial hand - wikipedia learned how to structure a community-edited reference site by doing, we can learn from them and exercise more control over policies and choices from the start to make the learning curve better

How we are different from Wikipedia

• Our standard of "notability" is much broader. Minor leaguers who played a handful of games in 1910 get a page in our Encyclopedia, but would not meet the Wikipedia standards for inclusion
• We will likely be more open to lists and tables than Wikipedia. Most minor league players' biographies will consist of little more than a list of teams played for, and their statistics. Bullet-point lists of facts will also be more welcomed than at Wikipedia, which prefers prose, although we are certainly hoping to have a large number of articles with nontrivial amounts of prose.
• Ours is an encyclopedia as much for researchers as it is for the general public. We may include material which is more "technical" in nature than Wikipedia would; for instance, a collection of notes on outstanding statistical discrepancies from a minor league's season, or provisional information (suitably indicated) which is in need of further research.

Our Model

• A publicly viewable wiki that is editable by all members and some others
  • Going to any page on this wiki will display a version of an article which has been approved by an "Editor"
  • Because editing will be restricted to registered persons, we anticipate malicious vandalism will not be a serious problem. Editors will mostly be responsible for maintaining a coherent set of conventions across the Encyclopedia, especially with regards to semantic markup, and keep an eye on certain parts of pages which are "expected" to be present (e.g., transclusions of statistical tables from other wikis). The editors may also encourage authors to include sources in their edits.
  • We believe this editorial model is viable. The function we envision is similar to the moderator function on SABR-L, which has been successfully staffed for over a decade. With an adequate group of editors, "approval" of submitted edits should have fairly fast turnaround, since reviewing edited pages is something that can be done a few pages at a time, fitting conveniently into real-life schedules. It is unlikely that the volume of edits would ever exceed the abilities of a group of editors to keep up; if many edits are being made, it is likely that the bulk of them are being submitted by one or a small number of contributors. In that case, those contributors likely would be invited to take on an editorial role, obviating the need for independent review of their contributions.
• Other, "back-end" related sites for the maintenance of particular datasets and project, including
  • Multiple additional wikis which most will never visit, which are publicly readable and writable by very few (1-10, probably) members in the leadership of a research committee
  • Relational databases for indexing and maintaining the "authoritative" content (e.g., the minor leagues database for performance statistics)
  • Most will never visit these because the content from the articles will be included in pages on the main wiki
• All sites integrated with one login/password
• The guidelines for article style will focus on making content searchable - if not when it's entered, then by an editor later
  • This means both text-searchable (which is easy and automatic) and data searching (which will be accomplished with appropriate use of semantic tags)
  • This may not mean there will be perfect searching from day 1, just that the content will be searchable. With properly indexed and tagged content, it will be easier to create good search tools later.