View source for Forum:Questreq data module

{{Forumheader|Redwood Grove|name=Questreq data module|subject=How to handle usage of "Required for completing" section under quest rewards section|notes=|closure=allow|archive=true|date=4 February 2022|user=Spineweilder|type=technical}}

<!-- Please put your content under this paragraph.  Be sure to sign your edits with four tildes ~᠎~ ~᠎~ -->
Hello esteemed editors,

Recently there has been a push to use [[Template:Required for completing]] on quest pages to show a list under the quest rewards of the quests that become unlocked ([[Special:Diff/14229623|example edit]]).

This template works by storing quest dependency data in [[Module:Questreq/data]] and automatically generating a list of unlocked quests on each page. Changing the output of the template is done by adjusting the data module or the logic in [[Module:Required for completing]].

Do we want to use this template, or should we stick with writing the unlocks list in regular wikitext?

==Discussion==
'''Regular wikitext''' - I'm worried the way this template hides the content in a data module makes it hard for editors to understand and edit. New editors won't be able to make corrections on the page and are unlikely to know how to navigate and correctly change the module. There's no good way to customize individual entries, such as details of diary tasks or other partial unlocks.

I think the benefit of the module sounds good at first, since we only have to store the quest depencies once and all the pages update for free. However, this is actually a much bigger maintenance burden than using normal wikitext, since we limit the number of people who can edit the page. Would we want to add special handling on the module to allow adding notes on individual entries? At that point we're asking editors to understand a new template interface, when I would rather have everything written on the page in obvious wikitext.

Additionally, this template is placed on quest pages that don't have any unlocks ([[Special:Diff/14229619|example]]), which I don't think is really useful to a reader. I think we should just omit the template in this case. [[User:Riblet15|Riblet15]] ([[User talk:Riblet15|talk]]) 02:19, 18 January 2022 (UTC)

'''Probably wikitext''' - This seems to be causing issues - the message on [[Talk:Tears_of_Guthix]] is kind of a major red flag that this isn't accessible to editors, and is prone to errors. {{Signatures/Cook Me Plox}} 02:23, 18 January 2022 (UTC)

'''Wikitext''' - Until there is an ''input'' on a page that will store in smw or something, and not a data page like there is now, then it should be wikitext for ease of editing. IF, hypothetically, the requirements section of [[Lumbridge & Draynor Diary#Hard]] caused quest requirements to be added for the info to be pulled from then I wouldn't mind it as much. But hard coding is a no-go. I also don't think my hypothetical will be an easy thing to accomplish as well, but I'd rather have that compared to a data module. {{User:Fjara/Signature}} 02:29, 18 January 2022 (UTC)

'''Comment''' - I'm fine either way for the listing of the direct requirements. However, I do like how the template can pull a list of all the things it's indirectly required for, so I think we should at least keep the data module around for that. Perhaps the addition of a scope parameter that can be set to "direct" or "indirect", with it being omitted causing it to display both.

Another thing that was mentioned is that it doesn't make note of things that just need the quest started (like anything that needs you to talk to a ghost needing Restless Ghost started). However, the data module already has things that just need it started listed as Started:, but for some reason the templates just strip the text without making mention of it. -- [[User:F-Lambda|F-Lambda]] ([[User talk:F-Lambda|talk]]) 04:08, 18 January 2022 (UTC)
:I'm not sure that showing indirect unlocks is useful to readers on the quest page. If someone wants to do one of the indirect quests they would end up looking at the page for the direct requirement first anyway. Anyone feel strongly that indirect requirements is actually useful to show? [[User:Riblet15|Riblet15]] ([[User talk:Riblet15|talk]]) 03:43, 19 January 2022 (UTC)
:: I don't think anything past the next immediate step is particularly useful. The reason it's hidden by default is because it isn't deemed worth it for the space it takes up. It makes much more sense to let people utilize [[RS:UCS]] to determine what should be listed there, if anything other than direct requirements. - [[User:Andmcadams|Andmcadams]] ([[User talk:Andmcadams|talk]]) 08:42, 22 January 2022 (UTC)

'''Wikitext''' - Per Fjara. --{{User:Legaia 2 Pla/Signature}} 15:30, 20 January 2022 (UTC)

'''Wikitext''' - I think putting this in a module is a bad call. Random editors don't generally know to edit modules, as can be seen on the talk page Cook linked and edits like this [https://oldschool.runescape.wiki/w/Ratcatchers?diff=14234840&oldid=14229785 here]. Like I mentioned on [[User talk:F-Lambda|F-Lambda's talk page]], this was a direction that RSW went, and I think overall it is a negative. It doesn't make things particularly easy to change and obscures the content, making it hard for editors to change. Breaking that module while updating the reqs for one quest would also potentially break all quest pages, though it seems unlikely to happen. The use of a module here also leads to very verbose sections, which may contain text when nothing requires the quest (as riblet said) or contain tons of duplicated text/info like on [[Barbarian_Training#Required_for_completing]] that make it hard to read. I have similar issues with the {{Required for completing}} template. This also completely defeats the purpose of the Boostable template by putting it in the data module, it is meant to be easy for random people to edit (all they need to do is add yes or no). I would strongly recommend avoiding this route in favor of doing some sort of pass to make things consistent if that was the original goal. - [[User:Andmcadams|Andmcadams]] ([[User talk:Andmcadams|talk]]) 08:42, 22 January 2022 (UTC)

'''Wikitext''' - I prefer wikitext. {{User:Crow/Signature}} 01:34, 26 January 2022 (UTC)

'''Module''' — I prefer module. I am in favour of making broad use of automation (templates/modules/transclusions) within the wiki, usually wherever possible. As with all MediaWiki projects, we are a [https://meta.wikimedia.org/wiki/Wikipedia_is_not_paper digital encyclopedia] and can, and '''should''' leverage the tools that are accessible to this project. It is almost always a net positive, as information can be quickly curated in one spot and all relevant fragments scattered across the wiki need not be hunted down — they are seamlessly updated. This also means reduced clutter in the RC queue, and far less chance that errors will be left unnoticed. Wikitext is obviously a good choice due to it being highly accessible... but a module is a far cleaner solution in my opinion. In theory, the module shouldn't need to be updated too frequently once the database has been compiled. If editors come across missing requirements that they are unable to update, they are generally knowledgeable enough to make their way to the Talk page. If and when the module needs to be updated, then it can be dealt with by an experienced editor as this will not be an everyday occurrence. Another option could include using a markup list to track requirements, which can then be parsed by the module on page load. This markup page would be easily accessible to an inexperienced editor, and could avoid needing to edit the module directly. This way, the inner mechanisms of the module won't be fiddled with (most users shouldn't need to do this anyway), and editors that do choose to will presumably be experienced with Lua. {{User:Owl/Signature}} 05:33, 26 January 2022 (UTC)
:I strongly disagree that adding technical complexity to the wiki is a net positive. The biggest asset of the wiki isn't the extensive MediaWiki tooling ecosystem, it's the 300,000 daily readers who can notice and instantly fix mistakes. Errors are far more likely to persist when we limit the potential editors down to the 10s of people who understand modules. We've already had 2 talk page reports of errors (with one still yet to be fixed by an experienced editor) - and who knows how many more who tried to edit, but didn't bother or know to leave a talk page message? There's really no way the benefits of a module here are worth the added complexity and toil incurred upon experienced editors. [[User:Riblet15|Riblet15]] ([[User talk:Riblet15|talk]]) 09:31, 28 January 2022 (UTC)

'''Wikitext''' - Adding onto what everyone else is saying, there's just been another comment by an editor who doesn't know how to edit it. (See [https://oldschool.runescape.wiki/w/Talk:Tai_Bwo_Wannai_Trio?diff=prev&oldid=14237412 here]) The module currently also seems to be sorting alphabetically (feel free to correct me on this) which doesn't always bode well. We'd want things such as diaries to be paired together. {{User:Kelsey/Signature}} 19:30, 26 January 2022 (UTC)

'''Closure''' - We shall stick to Wikitext. -- {{Signatures/Spineweilder}} 04:11, 4 February 2022 (UTC)