Bluesoul's Testing Page

This is for one-off testing of formatting on pages that don't look right in other sandboxes.

The Problem: We lack an organized system for discovering related content beyond objective tags. "Related content" in this context could be thematically similar, or simply an "If you like X, you'll like Y" sort of relation. If one wanted a list of every mouse-like or rat-like SCP, they can use the murine tag, but if a user wanted to get a list of all articles that involve an end-of-the-world scenario, or interviews with a god, or the exploits of the Foundation's IT department, or anything else outside the scope of those objective tags, that user is frankly up a creek. This hole in the discovery process has a knock-on effect that once an article, particularly a tale, falls off of the "Most Recently Created" list, the visibility is tremendously reduced.

The tag system solves other problems, but not this problem. Tagging currently deals with objective qualities that an article possesses. The Technical Team has no collective ambitions of being the arbiters of subjective qualities as well.

The scope of the affected content is significant. There are currently over 2700 tales on the wiki and of course a little over 4000 skips. Asking staff to shoulder the burden of improving discovery is a significant ask. A proposal involving tagging tales by genre involves a lot of work for a very unclear payoff, given that probably half of those 2700 tales would get lumped into horror.

A Possible Solution: I'd like to explore the idea of collections. At it's most basic, the idea is embodied in the User Curated Lists page.

A collection is a list of related articles, curated by one or more users for the purpose of aiding others in discovering content they would enjoy.

Now, a possible solution would be that we just really push the usage of the user-curated lists page, and it could potentially do everything I lay out below, but that page may quickly grow unwieldy in size. So this is my current proposed vision for the collection system.

  • A collection is created only in the collection: category/namespace.
  • Collections may be curated by:
    • A single user,
    • As a collaborative effort, or
    • By staff as a collaborative effort.
  • Collections are of one of three types:
    • Personal collections are curated lists in the style of a "favorites" or "best of" page. A user is limited to one personal collection page (there is a hard limit of 200,000 characters per page, this would take some time to fill).
    • Theme collections are lists that have something in common other than a genre. It could be things like articles that mention the Foundation's space program, or every tale with a strong female lead. Theme collections can be as vague or specific as needed. A user may create and/or contribute to any number of theme collections.
    • Genre collections are more self-explanatory, but can also be as vague or specific as desired. A user may create and/or contribute to any number of genre collections.
  • The requirements for creating or contributing to a collection are simply being a site member. I chose this because I don't believe the ability to curate a nice collection is dependent on someone's ability as an author, and vice-versa.
  • Any user may create a new collection, however…
    • Redundant collections are not allowed. Staff are tasked with making sure the collection is not a duplicate of an existing one.
    • Collections that may be perceived as inciting conflict or drama, such as "Bad dialogue", or "Good premise, poor execution" or even "awful premise but great execution" should not be allowed, authors will object to their works being present and for the purposes of content discovery, those collections buy us very little.
    • Standards for quality still exist for collections, including rating and deletion guidelines. Rating is relevant as a barometer for how useful a collection actually is, or how good a portrayal it is of a theme or genre.
    • A collection is not intended to substitute for an author page, though an author could use their own articles as jumping-off points for related content from others that they enjoy.
  • The only significant departure from the normal site rules and processes would be that author provenance does not apply to theme or genre collections. This is in an effort to reduce duplicate collections. In the event of an edit war, where the two parties cannot come to an agreement, the proposed edit goes up the Community Outreach chain with the CO captain having final say in the matter.
  • A collection may contain other collections. I bring this up specifically because I'd like to see a staff- or system-maintained list of personal collections by authors with over 3000 net upvotes, but also because I can't think of a good reason not to allow it.
  • Staff-curated collections are intended as a set of individual recommendations by staff members, i.e., "Staff Member X likes this article, Staff Member Y likes this other article." This is in contract to a "Royal We" list, i.e., "The Staff™ like all these articles." An example in another format would be Pitchfork's staff picks of music to be productive to. I envision this used for best of the quarter/year/contest picks mostly.

Collections would be primarily under the auspices of Community Outreach. Technical will need to maintain a small set of tags that are easy to use:

  • collection - Applied to all collection pages.
  • personal - Personal Collections
  • theme - Theme Collections
  • genre - Genre Collections
  • staff-curated - Staff-Curated Collections

The system will not require staff to review the inclusion of an article to the list, except in the cases that differing opinions lead to an edit war. The system will not require any changes to charter that I can determine.

Promoting Collections: Launching collections would benefit from a little publicity and planning. All Operational Staff and above are strongly encouraged to submit a personal collection and collaborate on at least 5 theme collections and 5 genre collections. There will be a 30-day moratorium on non-staff creating new collections so we can set the bar high for the future. I'd like to reach out to noteworthy authors and site members during that time and ask them if they're interested in helping launch the system with their personal collections and perhaps collaborating on starting some theme and genre collections.

A Collections link would feature in the Library section on the side and top bars.

As part of the normal front-page updates, we would include a featured collection.

The collections hub would allow for a sentence or two to explain the collection, if desired. No hard limit, but keep it to a sentence or two.

Other Notes: As for the coexistence of collections and tags, I want to reiterate the status quo of tags. All tags deal with objective qualities of an article. This is how it can function under the sole purview of Tech Team; asking us to be the arbiters of subjective qualities will only foster bad relations between Tech and the public. This proposal does not require any changes be made to the current way of running things, and is broadly compatible with Tech's aims and proposals regarding tagging. This coexistence is necessary if we want to take on the business of organizing subjective qualities and not cause a lot of conflict among staff and the users, and I think this is the most equitable way to go about the assignment of subjective qualities available to us.

Improving the visibility of tales is, to me, a worthwhile endeavor. Our most varied and often lengthiest articles are in a format that tend to quickly disappear from the public eye. The problem was slow to accumulate, and it will be slow to fix, but I'm confident that the rate of discovery will outpace the rate of article creation. I believe our content creators, youtubers, podcasters and such, will have a vested interest in maintaining quality collections, and really delving into the back catalog in search of those "deep cuts" that will really impress the audience. Moreover I think the collective memory of the community is quite good, but it hasn't been given a chance to prove it. The ability to add a link to a themed collection is something that practically every wiki member would possess, even those that would hesitate to write their own articles. I harbor no illusions that it will solve any problem quickly, but I do believe that it is a low-risk, low-overhead system that will immediately be a step in the right direction and will only increase in value with time.

At this time I'd like to request questions and comments on the concept and execution of the collections system.