Dragalia Lost Wiki:Discussion on Potential Fork

With the announcement of Gamepedia domains migrating to Fandom domains in early 2021, the administrators of the Dragalia Lost Wiki want to have a discussion with the community on whether or not to fork the Wiki.

To put it frankly, most of the admin team has long been wary of Fandom, which is why we chose Gamepedia as our Wiki host in the first place. We stuck around for the most part because we were told that things would get better, so the admin team gave this new Gamepedia/Fandom merger a chance.

However, with each change and announcement, we've been feeling less and less sure of the future if we stay.

• The initial release of their new editing UI called UCP is clunky. There is a redesign, but we have little faith in future changes and updates for it.
• Fandom's "customization" of extensions to match their platform can delay updating to the most recent versions of the extensions and also introduces potentially unwanted extension behavior.
• Mass homogenization of Wikis, making Wikis less flexible and customizable.
• Fandom's abundance of ads, especially those unrelated to the Wiki -- more akin to Buzzfeed-like ads.
• Fandom's poor communication about decisions like the Gamepedia → Fandom domain migration in such a short timeframe.
• We, first and foremost, see ourselves as a Wiki here to deliver information -- not a "fandom".

A couple of our editors are also part of communities where other Wikis have already made the decision to fork, so we find ourselves at a crossroad about whether to stay and see what Fandom brings or moving elsewhere.

How a Transition Would Work

Currently, we are undecided between two options if we do decide to fork:

• Hosting our Wiki on the Miraheze platform
• Independently Self-Hosting

In both cases, we would transition the Wiki to the domain "dragalialost.wiki" which the admin Elaeagnifolia already owns and currently redirects to this Wiki. Both options also offer a greater level of freedom in regards to flexibility and customizations of the Wiki compared to Fandom. This includes things like using different Wiki skins, having finer control over the mobile experience, etc.

The downsides of transitioning would include things like rebuilding the new Wiki's SEO and moving editors over to the new platform. Even if a consensus is reached to move, there will be some users who will not want to leave, which is understandable.

Miraheze

Miraheze is a non-profit volunteer group that hosts ad-free Wikis for free. All funds Miraheze receives are donation-based and go towards running Miraheze's servers.

Compared to self-hosting, the benefit is that the Dragalia staff and editors can continue to focus on content rather than concerning ourselves with also running the backend (Hardware, MediaWiki and extension upgrades, backend security, etc.).

On the flip side though, Miraheze is a volunteer-run group, servicing several hundred Wikis. It is not possible to have business-level service; however, based on experience, they are still quite knowledgeable and helpful.

There are also additional concerns though including:

• We're still beholden to Miraheze's "rulebook". For example, if an extension we want to use is denied by Miraheze, then that's that. Besides this though and other backend matters, they are usually pretty hands-off with how Wikis are ran.
• Miraheze's Financial Situation - Due to Miraheze solely depending on donations to fund the servers, there's some concern when donations aren't stable. For now though, the funds should still last for the foreseeable future, but it is something to keep in mind.
• Miraheze Performance - Some of the staff have worked on Miraheze Wikis before, and pages can potentially load slow.
• Miraheze Bugs - Bugs do get introduced from time-to-time as the Miraheze volunteer staff make changes, which may otherwise be caught when compared to staying with a company with paid IT.

Self-Hosting

Our other option is independently hosting the Wiki. This option would take the most effort to transition, as well as the most maintenance. However, this route basically gives us freedom to run the Wiki how we see fit, as there would be no upper management making decisions that we may or may not agree with. We would have full flexibility in managing the server resource-wise, extension-wise, code-wise, etc., but this also means all the responsibility of those also falls on the staff.

Some of the Dragalia staff do have technical backgrounds, but there may still be growing pains the first couple of months trying to fine-tune the backend server and other technical details.

Financially, server funds would be paid for by the current administrator Elaeagnifolia out-of-pocket. A couple of the other staff would also pitch in the first few months, and we would also look into opening donations in the long-run. We have no plans to add any form of ads outside donation links if we do self-host.

Going this route also gives us the potential opportunity to join NIWA (Nintendo Independent Wiki Alliance).

Pros and Cons Summary

The following is a list of the pros and cons of the possible different options:

Staying on Fandom Pros/Cons

Pros

• Fandom maintains backend so that we don't have to.
• Active paid staff and support for any issues we might not be able to resolve on our own.
• Won't have two versions of the Wiki, Wiki community stays together.
• No need to rebuild SEO compared to forking.

Cons

• Little flexibility and customization, as Fandom wants to keep the stylings between all their Wikis homogeneous.
• Ads, which mostly can be blocked, but they can still sometimes be bothersome especially on mobile.
• Fandom has strict guidelines to blocking ads site-wide, so to give any instruction on this, we'd have to manually write out a guide, instead of just installing ad-blocking themes wiki-wide. Note that not just ads, but also clunky layout, would be here to stay, without manual customizations from each individual user.
• Fandom has a manual Javascript review process, which could slow down potential customizations.
• Fandom's new UCP is clunky; while there is a redesign in the works, it's yet unknown whether it would fit a power-editor's needs
• Because of Fandom having its own platform, extensions might not always work as expected/are sometimes customized to match Fandom's platform. This may delay extension updates that would otherwise work on base MediaWiki.
• Some extensions which help QoL for editors are simply not allowed on Fandom.
• The brand change to Fandom from Gamepedia may come across as a bit "unprofessional" as a Wiki.

Miraheze Fork Pros/Cons

Pros

• More flexibility and customization. Miraheze is usually pretty hands-off with how Wikis are ran.
• Ability to better tweak the mobile experience.
• No ads.
• There will still be staff (although volunteer staff) to handle the backend so that we won't have to.

Cons

• Need to rebuild SEO and site visibility.
• Fandom does not allow deletion of Wikis, so two Wikis would exist.
• No backend control, so we're still somewhat beholden to Miraheze on decisions like adding extensions, backend setup, etc.
• Potential page load performance concerns.
• Bugs do occur from time-to-time, some which may not happen if we stay with Fandom.
• Miraheze funds are donation-based, so potential concerns about Miraheze's financial situation in the longterm.

Self-Host Fork Pros/Cons

Pros

• Full flexibility and customization (Extensions, styling, mobile experience, etc.).
• No ads planned. Elaeagnifolia will pay out-of-pocket with plans to open donation support. This may change if their financial situation changes, but this is not expected to happen in the foreseeable future.
• Not beholden to any third-party organization decisions. Any decision for the Wiki will be one made by the Wiki's own staff and community.
• More backend control.
• Potential to join NIWA (Nintendo Independent Wiki Alliance).

Cons

• Need to rebuild SEO and site visibility.
• Fandom does not allow deletion of Wikis, so two Wikis would exist.
• Dragalia Wiki Staff have to take on responsibility of also handling the backend (technically and financially).
• Potential hiccups at the start while finetuning/configuring servers.

Votes and Discussion

For a read on how Wikipedia discussion and voting works, see Wikipedia:Consensus.

The following templates are available to express your support, opposition, or if you want to abstain/comment.

• Support - Template:Support
• Oppose - Template:Oppose
•  Abstain - Template:Abstain
• ~~~~ - Signs post

Discussion is planned to close Dec. 1, 2020. A 2-week extension may be granted on a rolling basis if discussion is active and on-going.

Option 1: Stay on Fandom

• Oppose I have been providing updates in the Dragalia Lost Discord regarding UCP for the past several months. At this point in time, the pros of staying with UCP are vastly outweighed by the cons. I wish the UCP platform the best as it wraps up Phase 1 migrations and moves to the design in Phase 2. Canim44 (talk) 01:29, 17 November 2020 (UTC)
• Oppose While I'm unhappy with the changes Fandom is taking, I feel my decision is based more on the direction of our Wiki vs. the direction Fandom is taking simply diverging. I don't think they will be changing their policies because 1) They are a company running a business, so some of their decisions will no doubt center around that. 2) Changes seem to be targeting the casual user majority and to make Wiki editing easier/more accessible to the masses.

A lot of veteran editors have a "Don't change what isn't broke." mentality, and unfortunately, I believe most of the main editors on our Wiki are in the veteran editor camp. Most of the editors I know of here already have a plethora of experience editing Wikis, are proficient in HTML/CSS/JS, and don't need to be handheld by Fandom's changes -- changes which are helpful to the majority but serve as gates for more experienced editors instead. I do have some personal qualms as well, like with ads, with the Fandom name, etc. but I think it plays a lesser part compared to my primary statement. At the same time, I want to keep an open mind to the rest of the community's opinions as well. If the community overall decides to stay on Fandom, I will not be opposed to it. Elaeagnifolia (talk) 02:04, 17 November 2020 (UTC)

• Strongest oppose Fandom is just ad-bloat, has an awful mobile view, an awful editor, and the staff really only cares about making money off of the editors' work. They don't really care about the content on the wikis they host, only that it makes them money. I dislike them on principle, but also, because they don't hold the same core values for what's good for a wiki. They actually held a poll for Gamepedia admins, and they specifically asked in that poll, how important the Gamepedia branding was to us, and if we'd want to migrate elsewhere, should the Gamepedia branding be lost, in exchange for the Fandom branding. As per my vote in that poll, I vote strongly to leave Fandom. They saw the results of their own poll, from their own power users, and this is the result which I believe was coming to them. Accessibility and user experience on the Fandom platform has never been very good, and the editing experience has always been sub-par. I see no reason this would improve enough to warrant staying. As far as I'm concerned, they've made their bed, and need to sleep in it. Discoomastar (talk) 02:07, 17 November 2020 (UTC)
• Oppose I think it's probably in the best interests that the wiki should leave Fandom due to both technical (of which the browsing and editing part is what I'm more concerned with, especially since I browse wikis on mobile fairly often, and occasionally edit on the spot) and, to a lesser extent, policy issues (mostly with how the DL wiki should be versus the implications of staying within Fandom, which have already been explained in the above opinions from other users). LegendarySilke (talk) 11:43, 30 November 2020 (UTC)

Option 2: Move to Miraheze

• Support. I do have some concerns over the long-term financial stability and potential maintenance turnaround downsides, but given that Elaeagnifolia has been hosting some wikis there for some time and still feels it is worth recommending, and also given that I naively don't think it would be monumentally difficult to migrate further into self-hosting (at the same domain name anyway) if it doesn't work out (would take work, yes, but given various batch import tools options hopefully not insurmountable), I feel the most comfy with this option. As far as SEO and reader visibility goes, yeah, that'll continue to be a problem, but in a sense we do already have 2 sources of Dragalia information if taking into account Gamepress, and I do hold out hope that it's likely we at least won't have severe editorship fractures in our current group if a majority consensus is reached. -eave 00:25, 17 November 2020 (UTC)
•  Abstain I am fairly neutral on Miraheze. It works when it works. When it does not work, it can be more frustrating than current UCP outages, but the volunteer community is fairly regular with updates. Canim44 (talk) 01:29, 17 November 2020 (UTC)
• Weak support While as eave mentioned that I do have some Wikis on Miraheze, and some of them I'm quite happy with how they turned out, I will say that there have been issues on Miraheze, which Canim has mentioned can be even more frustrating than some of the problems we have on Fandom. I also do have concerns about Miraheze's financial situation and performance, but this option is free and does let us stay unconcerned with the backend details we'd otherwise have to worry about with self-hosting. I think it's also worth noting that Miraheze is providing a free service run by a volunteer staff, so I guess some concessions are necessary...and their staff is undoubtedly still very knowledgeable and try their best to remedy problems that crop up. If we do have to fork, I think self-hosting may be a better avenue, but this is the safer of the two options. Elaeagnifolia (talk) 02:04, 17 November 2020 (UTC)
• Weak support While I prefer the option of self hosting, I think Miraheze is still a viable alternative. Anything that's ad-free is better than with ads, for the end-user. And we'd still be afforded a reasonable amount of control. I just worry about their own issues with uptime, especially when we don't have any control over it. And I think in the event that they go under, or are acquired, or anything else that may cause us to want or need to move again, it'd be just this same deal all over again. But in general, I think it's an acceptable choice. Discoomastar (talk) 02:07, 17 November 2020 (UTC)
•  Abstain This is actually the first time I've ever heard of Miraheze (which may or may not say things about them), hence due to total unfamiliarity, I cannot provide any comment on going with Miraheze. LegendarySilke (talk) 11:38, 30 November 2020 (UTC)

Option 3: Move to Self-Hosting

• Support I support this option for two main reasons:
  1. Ability for wiki admins to truly control the back-end
  2. Extensibility of the platform if the current data outgrows cargo's capabilities Canim44 (talk) 01:29, 17 November 2020 (UTC)
• Support I'm split between this and Option 2, but lean more towards this option. While this option is basically the most risky as it requires financial investment as well as needing staff to step up to the plate to handle backend server maintenance, it gives us the most freedom and complete say on how to run the Wiki. This includes things like backend control, fine-tuning issues that may appear (e.g. Optimizing performance, mobile design, etc.) as well as the ability to expand if needed. While it is an additional responsibility and there may be a rocky start in configuring the Wiki, I do think in the longterm we have the personnel and staff with enough experience to make the jump and to make the new Wiki just as good, if not better, than the original. Elaeagnifolia (talk) 02:04, 17 November 2020 (UTC)
• Strongest support I am in favor of self hosting, due to the amount of control it would allow us, and to avoid any potential future acquisitions or buyouts which may lead to a similar issue in the future. We've seen how some other wikis moved away from Fandom to Gamepedia years ago, only to now be forced right back, because they're still tied to a third party which could become acquired at any time. Additionally, as someone who has technical knowledge, as well as experience with self hosting a wiki, I'd be able to assist in that regard. I can also mention that I've been much less active an editor here since the removal of the CSS extension -- but we wouldn't have to worry about such things being pulled out from us, should we self-host, as we'd be in full control. Discoomastar (talk) 02:07, 17 November 2020 (UTC)
• Support I personally think it'd be better for most of the editors if we were to move it sooner than later. Fandom can get messy to handle with and deal at times (from working with some pages by Fandom once) and even though there will no longer be a support to help out as easily, having the flexibility to modify pages as desired is quite nice. I know it's already mentioned that Elae is managing the hosting but if you need extra hosting power I can help. Additionally, the possibility of connecting with NIWA could potentially bring some new faces to the wiki, which could help out. Admittedly, I'm not experienced with developing things but I can fine tune some of the info on pages when/if needed. MIAV-Volca23 05:20, 17 November 2020 (UTC)
• Support Fully controlling the future of your own wiki should result in great things with good governance. It also removes most external factors. The ride might be rough early on, but I believe that the wiki will be better off with benefits that going self-hosted can bring. LegendarySilke (talk) 12:43, 30 November 2020 (UTC)

General comments

• Regarding Fandom having a JS review process, yes this is certainly true and it's one of the things that chafes the most about it for me (though I understand it's so wikis can't just remove ads on their own/do anti-TOS customizations), but I'm not sure if they've indicated that this will necessarily be the case for GP wikis being moved over? If they have, then I strongly support moving away! -eave 00:25, 17 November 2020 (UTC)
• A Reddit thread was created on 16 Nov 2020. Canim44 (talk) 03:29, 21 November 2020 (UTC)