After 30 November 2020, updates for my addons are unlikely to appear on CurseForge.
CurseForge has been acquired by Overwolf. The site itself is scheduled to go into a read-only mode on November 30, 2020, at which point the personal and account information transfer is likely to be finalized. For residents of the European Economic Area, Overwolf asserts that this transfer requires user consent. Overwolf intends to continue distributing addons already published on CurseForge regardless, while removing attribution from addons whose authors do not consent to the transfer. I have not consented to this data transfer, and therefore expect to not be able to publish further updates on CurseForge after the transition.
There are three reasons for saying goodbye at this point:
- Twitch's CurseForge has practically been abandonware for a year.
- Overwolf has been terrible at managing the transition.
- User experience in a CurseForge monoculture will likely get worse.
Last year at CurseForge
Curse has spawned two asset-hosting platforms that have since been sold off: CurseForge and Gamepedia1. Both have gone through a period of expansion characterized by a desire to host everything, prioritizing visitor growth over the interest or ability to deliver a good experience on the created sites. In expanding platforms to support more and more sites, developer focus shifts from improving individual sites to improving the platform itself ("we added achievements, and twitch auth!").
Over the past year, CurseForge has basically had no one in a position to change how the site itself works. This leads to obvious problems: WildStar mods are still being featured on the front page (WildStar itself shut down in 2018); site CSS is still broken in Safari; the Twitch App still gets Classic and Modern WoW addons confused. Twitch's handling of user feedback through UserVoice, summarized by the recurring refrain of "We are Listening!", is just sad.
Overwolf itself is an ad-delivery platform for application developers, allowing monetization through in-game ads displayed in an overlay. The new CurseForge app is an application for managing addons and game mods made for the Overwolf platform. Rather than starting with an addon distribution site that happens to support itself through advertising, the new CurseForge started with an ad-delivery mechanism and built an app to manage addons in order to attract viewers. I wonder how the development resources will be prioritized between providing a good user experience and delivering the ads in a year's time.
Data transfer, and consent
The Twitch-to-Overwolf transition has been terribly communicated.
The current CurseForge site doesn't appear to acknowledge the
doom smooth transition due to occur next week. There is no overview of
what will happen when, and news is delivered via ad-hoc Twitter and Medium
posts, often days before the deadlines being announced. It is as if CurseForge
forgot it was a website, and that websites can tell their visitors about things.
The plan to strip attribution from existing addon pages (renaming some author accounts to Forge_User_[RandomNumber]) seems like flagrant violation of moral rights. If Overwolf's assertion that consent is required to maintain display of public, pseudonymous usernames is taken for granted, Overwolf should clearly not continue to distribute addon archives without acquiring that consent, as these archives most likely contain similar information about the author(s). If consent is not required to maintain attribution, stripping it seems coercive. Regardless, continuing to host old versions without author consent probably infringes copyright, as Twitch's license to do so terminates once these addons are deleted from "Twitch Services".
Overwolf seems to treat authors who take no action on the consent question as having consented to the transfer2. It is not clear how this could ever meet GDPR's standard for "freely given, specific, informed and unambiguous" consent. I can't even figure out what the data I'm supposed to consent to being transferred is.
There are other issues, including scheduling a period of limited moderation in the week following the Shadowlands launch (since postponed for existing addons following a community outcry). Nothing about this transition inspires confidence in anyone involved in it.
An uncertain future
Beyond increasing the install base for the Overwolf platform, it is unclear what Overwolf intends to do with CurseForge.
The primary business model, displaying ads to people playing video games, seems at odds with the interaction model of an efficient addon management app (which mostly should get out of your face as soon as possible). Given Blizzard's insistence that ads should not appear in the addons themselves, it seems unlikely that an in-game ad overlay would be acceptable. That leaves displaying advertising through the app itself. The current volume and style of advertising within the app is probably the least intrusive it will ever be — it is important not to alienate everyone on launch. Once the software is widely installed, knobs can be turned and verification cans will need to be drunk to continue.
Will CurseForge, the website, survive as a place where one can download addons with a web browser? There is certainly an incentive to push people towards the app, which mitigates both browser-based ad blockers and unauthorized downloads™ through third-party addon managers. For now, there are no changes announced to the site itself.
It bears acknowledging that distributing billions of mod files is not free, and that CurseForge pays addon developers, a practice that Overwolf intends to continue. Nevertheless, I value user experience more than the revenue from CurseForge, and do not like to be used as a convenient excuse to require you to install ad-serving software.
Based on how things have been going, I no longer want to use CurseForge as the primary distribution site for my addons. Depending on how the attribution/consent/account issues resolve, it may still be a useful secondary distribution site to reach users who find Overwolf's client/site pleasant enough.
Existing alternatives are somewhat unappealing, with nobody providing all of the desirable features in an addon site:
- Addon discovery: finding addons you didn't know you needed is nice.
- Development tooling: user-contributed localization management, issue trackers, wiki pages and/or forums are useful.
- Browsing experience: it would be nice to not be greeted by a cookie consent dialog on every uncookied visit. It would be nice to have ads the site administrators could stand behind, instead of blaming the ad networks.
- Active development and content moderation: generally, a good website requires human input, both in terms of keeping up with the web standards, implementing a new feature every few years, and enforcing standards for the quality of published content.
- Some mechanism for supporting addon authors. I don't think a Patreon/PayPal button for every author scales well enough: people run too many adds to support everyone each month, which causes friction in deciding to donate to any one project. A broader subscription + distribution option might make more sense, and could also support people whose work addon authors rely on (like the people editing API documentation).
I would, of course, also like a unicorn.
For now, I expect to publish addon updates here. Perhaps, over time, that will grow into a site that addresses the above concerns. And then, of course, spawn an addon management client. With blackjack. And unicorns.