Breaking Extensions so much

[AlterZwerg]

With Foundry, I was forced to lock the version when I got everything working for a campaign. I could absolutely guarantee and afternoon's SysAdmin work for myself if I ever wanted to update.

With FGU, I update weekly and if anything breaks, it'll be one of the small number of extensions which I just turn off for the week. And I rarely need to do that.

But at least in Foundry you have the option to keep stuff from updating by locking it or you can roll back to a previous (working) version.
This option is simply not a thing with FG but would be very helpful for issues like this.

I don´t run 5E and stopped shelling money out to WotC long ago. I want FG to improve so personally i can live very well with these growing pains.

[eporrini]

Extensions are a necessity for me and my players for 5e. I am not a developer, but it feels like we change core variables too often, which cause the syntax of extensions to require significant rework at times. My frustration lies in the FGU development team taking no accountability for the breaking of extensions that are critical to the overall use of FGU. Developers need to work in a less invasive way where possible to avoid levelling extensions during updates. The amount of impact recently has IMO far exceeded what I have seen in the past in terms of breaking/significant rework required.

I agree with some posters who said "don't update". That's sad, truly. I am going to be in a mode where I get stable and tell all of my players we will not be updating, missing out on new features and capabilities being introduced. I can only imagine that when I add new purchased content, at some point that will error out due to not updating...

My biggest issue presently is that certain key developers of extensions (BMOS, MeAndUnique to name a couple) don't have time or have stepped away from their extensions. I do not blame them, they have no obligation to provide updates to free extensions and it's a lot of work. When this happens, other developer "heroes" step in sometimes to help, but then you get into a Gitlab repo situation of hunting through threads for updates and knowing auto updates are out of the picture. There is no good way for a developer to take over abandoned extensions. I think there should be something in a EULA that says if an extension is abandoned another dev is allowed to lift it out and start updating it, even if under a new name or revision. I believe that certain abandoned (foundational) extensions could be taken by the FGU developers and incorporated. You likely have the stats on what's being pulled down from the Forge to gauge overall customer usage.

Bottom line, we need the FGU devs and leadership to get with extension devs to work through a better plan than exists today that makes sense for everyone. Extensions are too important (some are critical) to go through this suffrage each time we update.

[bwatford]

While I personally use and enjoy several extensions, none are essential for playing 5e or any other ruleset. The base functionality is sufficient for each ruleset. Extensions do make life easier by adding automation and other features, but they are not a necessity. We often become accustomed to the convenience they offer and forget that the game functions well without them, albeit with a bit more effort. It is still significantly better than pen and paper.

The developers provide a test channel for extension writers to work within before a release. The main issues are the lack of documentation from the developers and the limited time available to the authors. Every update tends to break something, and I do not foresee this changing. However, we can strive to minimize the impact.

I believe the development cycles are too rapid, almost weekly, which poses a significant challenge for developers who lack the documentation needed to comply in a timely manner. Since no one is making a living off extensions, free time is limited. These cycles should occur once a month at most, with weeks of lead time, rather than an active development cycle where changes are added to the test channel daily, impacting extensions.

Development cycles need better management, with documentation of changes provided as soon as they enter the test channel. Subsequent updates should only address fixes, not add new features. Changes should only enter the test channel once they are finalized on the development side.

While there is a rush to release new features, the current approach appears quite amateurish and ultimately drives users and extension writers away from the software. We have lost many talented contributors in the past 6-8 months due to the lack of time to continue development, leaving us with several “good” but unsupported extensions that are being patched together by volunteers.

We need to improve.

Phase 1: Developers work behind the scenes to develop the next version.

Phase 2: Documentation of upcoming changes is sent to third-party developers.

Phase 3: The version enters the testing phase. Extension authors work on updated versions during this period, which lasts for at least 30 days. No additional features are added during this time, except for bug fixes. The software developers collaborate with third-party developers to ensure everything is ready for the official release.

Phase 4: Official release.

[Zacchaeus]

There have only been three ruleset updates this year, in March, August and September.

Anything else has been bug fixes or updates to modules.

[Vass_Dts]

The amount of impact recently has IMO far exceeded what I have seen in the past in terms of breaking/significant rework required.

Fantasy Grounds is a 20-years-old program. It has been iteratively been build upon brick by brick. When the jump to Unity occurred, FG wasn't exactly re-written from the ground up. A colossal portion of its original code was just transplanted into Unity. During the first stage of this transition into Unity, the aim was to have the thing work in much the same way as it used to, but with some additional features whenever possible.

Now we are in the stage where the Dev team want to get rid of all bloated portions of the code, to streamline and optimize the software going forward. This endeavor is the primary reason you've seen more breaks than usual over the past year and a half or so. While some breaks in 5e did happen due to the 2024 implementation of the rules, many actually happened because of the streamlining of the base code. And while it is frustrating for us users (hey, I still haven't updated for instance), you can't have the software remain static. That's not good for any of us.

But yes, SW, from now on should be take steps to address possible breakages. Documentation is the key, as @bwatford said (among other things). Breaks will never stop happening as Ruin said earlier. The goal is to make them less frequent and less painful (particularly for extension devs).

EDIT/ADDENDUM: All of the above is my personal understanding of the situation, and I may have some details wrong, but the general gist applies.

[Laerun]

Fantasy Grounds Community Extensions:

Fantasy Grounds extensions are powerful tools developed by the community to enhance gameplay, customize rulesets, and automate tasks within the platform. However, these extensions come with their own set of challenges, particularly related to maintenance, compatibility, and ongoing support.

Key Insights on Fantasy Grounds Extensions:

Community-Driven Development: Extensions are typically created by individual developers who initially build them to improve their own gameplay. Over time, these tools are shared on platforms like the Fantasy Grounds Forge, where they can be distributed for free or at a low cost, giving them more visibility and allowing other players to benefit from these enhancements.

Technical Debt and Maintenance Challenges: Many extension developers face the reality of technical debt. As their portfolio of extensions grows, so does the difficulty of keeping them updated and compatible with the latest Fantasy Grounds updates. Frequent updates to Fantasy Grounds' core and specific rulesets can lead to breakages that developers must address. However, testing all possible scenarios is often beyond the capacity of individual developers, especially as they juggle multiple extensions.

Updating and Compatibility Issues: Extensions can conflict with each other or with updated rulesets, causing unexpected behavior during gameplay. Since SmiteWorks (the company behind Fantasy Grounds) provides indirect support but does not maintain these third-party extensions, the responsibility for updates and fixes lies primarily with the original developer. This can lead to delays in updates or, in some cases, developers abandoning their extensions due to the workload involved.

No Universal Testing Framework: There is no comprehensive testing system that ensures all extensions work flawlessly across various use cases and ruleset updates. Developers rely on feedback from users to identify and fix issues that arise, often only after an extension has already been impacted by an update.

Best Practices for Extension Developers: Developers are encouraged to keep an eye on beta announcements for Fantasy Grounds and to test their extensions on these versions in advance of major updates. By carefully managing changes and minimizing overrides of core elements, developers can reduce the frequency and impact of breakages. Creating backup copies of ruleset files they modify can also help streamline the update process when core changes occur.

User Awareness and Acceptance: For users, understanding the nature of extensions is crucial. While they add great value and convenience, they are not essential and come with inherent risks due to the evolving nature of Fantasy Grounds. It’s important for users to stay informed about the extensions they use and be prepared for potential disruptions during updates.

Extensions are a testament to the creativity and dedication of the Fantasy Grounds community, offering enhanced gameplay experiences. However, they also require ongoing commitment from both developers and users to manage and mitigate the challenges that come with integrating these custom tools into the evolving Fantasy Grounds ecosystem. For more information and guidance on using and developing extensions, you can visit the Fantasy Grounds Forge author's page where you subscribed to any given extension, and the Developer Guide Atlassian and/or these official FG forums.

https://fantasygroundsunity.atlassia...eveloper+Guide
https://forge.fantasygrounds.com/

[Mike Serfass]

As for testing, that can't be done by just the devs. Developers actually aren't the best people to QA their own code, much like a writer shouldn't be the editor of their own stories. There are reasons those jobs are separate.
We need volunteers from the community to switch to the test channel and test the CoreRPG changes, the ruleset changes, and the extensions.
As both a ruleset and extension dev, I've missed bugs because I just didn't have time to test everything, and nobody else tested that functionality, and it wasn't caught until after release.
Having test options doesn't help if nobody tests.
Volunteers willing to jump into the test channel are rare.
People willing to report and explain a bug are uncommon.
Those who criticize and complain about bugs without useful input are too common.

[ghen]

I think jharp's message about automatically unloading extensions means well, but a lot of extensions don't need constant updates so they'd just be waiting around for a version number increase before they could get used.

The best solution is better documentation when ruleset code is changed. Like, real nitty-gritty stuff. Effortless extensions support has GOT to be worth more to SW's bottom line then the extra time it takes to document changes as they are implemented.

Less time for extension devs chasing bugs means less burnout (deprecated extensions=dead platform), less fear making extensions work on more rulesets, and more time to make more extensions.

[BushViper]

I understand the complexities and nuances of both sides of the debate, but I'd like to point out two things:

1. The customer is always right. I don't mean that in the literal sense, but rather more broadly. If a company is not providing the products/services that the majority of the customers are asking for it doesn't matter what the company thinks because they're not going to remain in business for long. Right now, we're seeing failure after failure in the gaming industry as companies continue to tell its customers what they should want/like. We know what we want and we tell you. Often.

Ignore your customers at your peril.

2. Smiteworks created, promotes, and profits from the Forge. There's zero debate about SW having some responsibility in reducing the pain points of using extensions. When they work, extensions objectively improve the user experience. For some, that improvement is so massive that a stock client is a deal breaker. However, when maintaining an extension becomes a repeated source of irritation and inconvenience it can (and does) undermine customer loyalty and drive people away.

I'm not suggesting that the issue of extensions is a harbinger of doom, but it's definitely a very important element. And I don't think the community is asking for too much in their desire for extensions to be broken less often and more resources be spent absorbing the incredible ideas that already exist rather than be wasted on producing gimmicks.

Even the people that are more inclined to side with Smiteworks on this topic clearly understand the importance and value of extensions so their responses are half-hearted and more-or-less -- "You're right, but you don't have to be a douche about it." The latter is probably true, but the reality is that we live in a world where the customer has the expectation that the product they buy works and when it doesn't, it pisses people off. Even more so when the product is literally broken by the company that sold it to you to begin with.

[MartianXi]

You likely have the stats on what's being pulled down from the Forge to gauge overall customer usage.

This is a great point... I'm sure SW could pull data from the Forge correlated against a list of active users and get an idea of which extensions are used the most. If they aren't too robust they should be seen as valuable and maybe retooled and added to a ruleset.

I didn't even know about the suggestions voting page until last week and forums and voting systems only show a portion of reality. Those who are proactive will post and vote but they only represent a portion of users. Follow the trail of where the forge money is focused and you'll learn the truth.