DDO wiki talk:Structure guidelines

I've created a first draft of guidelines to be used when determining which item pages to include in the corresponding item effect category. This is based on what I was able to determine was the current state of the wiki, along with an attempt to define a simple set of criteria that could be followed.

I encourage editors to provide feedback on which criteria they feel should be used to determine category membership by adding your name in the appropriate cell of the below table. We will review and propose revised guidelines if needed based on the responses.

If you have ideas on what the proposed guidelines should be, or how to more streamline what I've created, please create a sandbox page for reference, and then provide link on this page for collaboration.


 * I just whipped this together, feel free to refactor/clean up/streamline to make the content more readable. (Or if anyone has questions or needs more clarity, but please review the Guideline page first to understand the terms used here.)
 * (I will expand later on why I take the position that any items that do not have the Enchantment persistently (directly) should not be categorized (in the existing structure) as a follow up, but wanted to get something put together here so we could start the process for getting feedback and working towards defining guidelines that were most popular.) &rArr; Joenuts (Contribs • Message • Email ) 06:36, October 31, 2020 (EDT)

Basic questions
I think the article fails to ask the basic questions. You ask "How should articles in the Item namespace and specifically those 6 "categories" of items should be categorized?". However, an answer would need to consider "What do we hope to gain by categorizing them one way or another?". And in order to answer that, we also need to find out "What are categories for? How do users and how do editors use categories?". Thankfully, we are not alone in this matter. Wikipedia has already asked those same questions and a lot more minds than us created a solution (which even in their case is not universal). So, to quote https://en.wikipedia.org/wiki/Wikipedia:Categorization, "The central goal of the category system is to provide navigational links to Wikipedia pages in a hierarchy of categories which readers, knowing essential—defining—characteristics of a topic, can browse and quickly find sets of pages on topics that are defined by those characteristics. Categories are not the only means of enabling users to browse sets of related pages. Other tools which may be used instead of or alongside categories in particular instances include lists and navigation boxes.". I agree with that, but we also need to consider any differences that our wiki may have compared to wikipedia concerning browsing of topics. Our wiki is more than just an encyclopedia where a topic gets explained and readers know the names of topics they are looking for. Wikipedia does not directly answer questions like "Which color is best to use when painting a house?". Our wiki does because it's also a guide to playing the game. When a user says "I am level 10 looking for more Strength and a weapon against demons. What should I look for?" our wiki tries to answer that question in the most direct way possible while not preventing other question combinations from having a somewhat direct answer as well. Some topics in our wiki also change more rapidly than topic in wikipedia. A quest or an item or an enhancement tree may change properties frequently and thus needing to change the "relevant mentions" in other pages. For example, if paladins can no longer turn undead, the turn undead page must also be changed to remove any mentions of paladins. If Shadow Crypt's xp is lowered then pages listing top xp quests may need to change. To achieve that result of displaying tables that can be sorted or listing relevant pages, our wiki also uses DPL (which used to be NCL). Because of the computational limitations of DPL, getting a relevant page list is easier if those pages are in the same category. For example, one could not use DPL to get all the monsters that don't have a picture because the list of pages that use Template:Monster are more than 500 which is the limit of DPL results. If those monster pages were categorized, then presto you can have a list. In that regard, we can say that our wiki needs 2 kinds of categories. 1 that users can use to navigate between relevant topics (although this rarely happens in my opinion) and 1 that editors use to create lists of pages using DPL. So, to come back to the original question: If an item page may promt a user to search for similar pages, then that item page should be in the categories of those meanings. If an item needs to belong to a table or a list because of its properties, then it needs to belong to those categories. Keep in mind though that you can't overuse the 1st type of categorization because of https://en.wikipedia.org/wiki/Wikipedia:Overcategorization. Having too many pages in a category with varying degrees of relevance is category clutter and leads to that category being unusable (unless it's supposed to be used by DPL to create easier to digest tables or lists). So, the category Category:Raw ingredients is not usable by a user (in addition to the fact that it contains historic items) but it may be usable by a DPL query searching for a list of ingredients related to a crafting system without the need to have those ingredients in a category "raw ingredients related to this crafting system". To answer to the examples you have provided in the main topic: Obviously items with persistent effects belong to those categories. Item pages that do not have the effects directly should be in the categories or not according to the following question: "Is there a page that features the same item WITH those effects?" For example, if Vibrant Purple Ioun Stone has a page called Vibrant Purple Ioun Stone (unsuppressed), then only the unsuppressed version should be in those categories. However, if we decided to merge those two pages into one that has 2 item templates in it (like we do with monsters that have the same name), then that merged page would have the categories. About crafted items like Green Steel or Slave Lords, we could not place them in the categories to reduce clutter but still have a mention of them in the lists we create. If that's not possible, then we should get them in the categories in order for them to appear in the lists. As for ingredients or items that turn into other items by crafting or turn-ins, then those don't belong in the categories or the lists because the final items are probably already there. To summarize, I don't think we can have a universal rule about item categorization except: "Will categorizing this page help or hinder the users and the editors?" &rArr; Faltout (Contribs • Message • Email ) 10:42, October 31, 2020 (EDT)


 * Faltout, I agree completely that answering the question "Will categorizing this page help or hinder the users and the editors?" to decide which pages should be included in the various categories and what structure to frame up. While I believe every editor you ask "should we include pages that are useful?" would answer a strong yes, users may disagree on how to define "useful" (this is what got us here). If you're suggesting that we use the "Is it useful?" question as a basis for determining inclusion, it would be helpful to provide some clarification on the following questions. (1) How do we decide on whether including a page is or is not useful? (2) At what level of granularity do we need to determine whether category inclusion is useful (page by page, or collections of pages)? (3) If we're not determining on a page by page basis, what criteria would you use to define collections of pages to then be able to more generally decide?
 * For your suggestion of a criteria defined as "Is there a page that features the same item with those effects?", this begs the question : How do we determine when unique pages should be created for items that have various levels of enchantments? (Never/Sometimes?/Editor(creator) discretion/Always). My thought process on this is a bit different. I think we START with identifying ways we can group items by some common concept, then we decide if those commonalities warrant categorization, and decide how to build the categorized based on what's determined. Whether or not a page exists already with the same effect is a bit confusing to me as a basis, as (I feel) we should be defining >>when to create<< pages, not letting the fact that pages exist or not determine >>when to categorize<<. &rArr; Joenuts (Contribs • Message • Email ) 15:54, October 31, 2020 (EDT)


 * That's why I also provided context for the meaning of "usefulness" (and why I had to write such a long post). "2 kinds of categories. 1 that users can use to navigate between relevant topics (although this rarely happens in my opinion) and 1 that editors use to create lists of pages using DPL.". The 1st kind of category is wikipedia style categories. I've linked wikipedia's entry on categorization to see what rules they provide to categorize pages. Basically it's something like " is " where category name is a "defining" quality (and if people disagree on what "defining" means then they sort it out on a per case basis). Because the 1st kind of categories is used for user search, they need to remain relatively small (by subcategorizing if they become too large) unless they are also a DPL category. The 2nd kind of category is DPL helper categories. While it's hard to restrict which articles end up there because they are usually automatically categorized, if an article is in such a category then it means the intention is to include it in a DPL search.
 * So, the rules should be: 1. Place each page in the defining categories and subcategories. Items in items, quests in quests, guides in guides, etc. 2. If you want the page to show up in some DPL table place it in the category for that DPL table. 3. If there is already a better duplicate of this page in some DPL table, then don't place that page in this category. (This answers "should we place both lavender stones in the category"). 4. If this article appears in a DPL table will it be helpfull to the user? For example, how is it helpful for a reader to have Kor Kaza runes show up in Quality Magnetism 48 items when a reader is searching for magetism items (and not what the Kor Kaza runes are)?
 * "How do we determine when unique pages should be created for items that have various levels of enchantments?" I think editor discretion. The game has multiple crafting systems/upgradable items/random items/etc. We can't possible make a rule for all of them. Can you imagine trying to create an item page for every possible greensteel combination? The goal is to include some kind of result in DPL tables for those systems in one way or another. So, if someone searches for a holy weapon at level 12, there should be some mention that greensteel has that. However, this is either very hard or impossible altogether. So, it's ok if we leave it to "whatever works".
 * "Whether or not a page exists already with the same effect is a bit confusing to me as a basis, as (I feel) we should be defining >>when to create<< pages, not letting the fact that pages exist or not determine >>when to categorize<<" So that's why my main drive is the end result. In the end, what I want to see is a page with that item appear in a table when I search for that effect. And I don't want 7 instances of that item appearing, although if it can't be helped it's not such a big deal. So I don't really care how we get to that point, just that we do. As for when to create pages, I'm going with editor discretion. The Cannith challenge gear probably deserves multiple pages for the different levels. However, each page could have 3 entries for the levels of upgrades. Or it could be one entry that includes all 3 upgrades. The main goals are to have the page show up when I search for the effects of one of the upgrades AND for the page itself to be readable (having 13 entries of Spare Hand in one page would make it extremelly long).
 * &rArr; Faltout (Contribs • Message • Email ) 17:52, October 31, 2020 (EDT)