Wasting time
I realized over the past year that I spend a lot of time on various tasks during my day.
Building Maven modules
Building Maven modules of ItemsAdder takes me around 30 seconds at the moment.
I bought my PC in 2015, and my i7-4790k might be nearing its end, which is the reason I decided to build a new PC with the most recent i9-14900KF.
This should definitely speed up the build time as I recompile my projects hundreds of times each day during development.
No. Hotreload is not an option due to how my Maven setup works. Lot of my NMS classes are dynamically loaded and hotreload always fails due to the custom hooks.
Another limitation of hotreload is the fact that new methods and new properties break the hotreload, so it's pretty useless at this stage, most of my values are set in the config.yml, so I can easily swap them. Hotreload is not useful for my projects.
Answering support requests
My current workflow to handle support requests
- check if the issue is readable and understandable
- check if it has every information needed for me to reproduce the issue on my machine
- setup a clean test server
- start the client (multiple if needed)
- setup ItemsAdder as the customer did
- try to reproduce the issue
This whole task requires from 30 to 1 hour, depends on the gravity of complexity of the issue.
If the issue is reproducible, then I proceed to check the cause of the issue.
The cause of the issue is bind to how the game works, how another plugin works, how my plugin works.
In each of these cases, I have to spend a lot of time documenting how to reproduce the issue to fix it and make sure it will never appear again.
In some cases I have to reimplement a fix for EVERY implementation/version of the server and for every client version (lot of servers are sadly using ViaVersion).
Servers using unstable stuff
A lot of servers are currently using solutions like ViaVersion+ViaBackwards and Geyser.
These hacks are a nightmare to handle because they make the gameplay unpredictable due to how each client handles stuff in a very different way. This causes confusion both on the customer and on the players due to how some glitches are introduced in the game.
My opinion is that there is no real advantage to continue supporting Geyser due to how different the Java client behaves compared to the Bedrock client.
Solution
I have no solution for now other than continuing to avoid supporting Geyser.
Invalid support requests
As I said previously in one of my old blog posts: I waste ton of time organizing support requests on Github.
Currently, 31% (1012/3270) of the Github issues pages of ItemsAdder are invalid (data obtained on 2024-Jan-23).
This requires me to actually provide an explanation to the customer, causing both frustration on my side and on theirs, because most of them don't accept their issue to be closed without an instant fix.
Solution
I have no solution for now other than continuing to handle reviews this way.
Bad reviews
Some customers are very stubborn and continue bumping threads even if the issue cannot be reproduced, they are running old versions of the plugin or server, they are using unofficial forks or even if they don't provide any information about the issue other than "it not work, pls help".
This results in 1 star reviews which have the only purpose of trying to get my attention and prioritize their support requests.
I spend half of the 2023 ignoring bad reviews instead of prioritizing these customers, as it's disrespectful both for me and for the other patient customers.
Instead, I'm concentrating on making my products better and helping customers which correctly provide information about their issue.
During this initial period of time I noticed a very large increase in the customers amount and, I'm sure the trust of customers is increased too.
Good reviewers will always come due to the quality of the plugin updates.
Solution
Ignore the bad lazy reviews and concentrate on making the plugin better.
License checking
Currently, BuiltByBit has no easy way for me to check if a user purchased a resource using their payment email, meanwhile PayPal and Stripe allows me to do that by scraping my own email notifications.
So for now, I am forced to manually verify purchased on my forum and on the Discord server for these customers.
Solution
The next thing to do is find a way to automate that also for people who are buying using BuiltByBit.
Lack of staff members
The sad reality of my business is that the staff members that I hire are always losing interest after a few weeks or during the first months.
This is something I had to face since 2019, the first time when I started hiring staff members.
This is a clear proof of how frustrating and hard is to deal with customers in this business type.
Solution
An idea I had during this year was to teach an AI how to answer the general support request questions, as most of them are generic questions who have answer in the plugin wiki/docs.
I'm not really into the idea of replacing humans with AI, but this is a very frustrating and repetitive task, so I don't see any issue on doing that.
Beta testing team lack of response
The beta testing team for my plugins is unresponsive, unmotivated, and most of the beta builds I send are not even downloaded.
This is a big issue since it vanishes completely the purpose of having a selected set of members who test my plugins to avoid the whole customers to experience issues.
Solution
Post beta builds on a public accessible place, so people can autonomously access the builds and test them.
I will obviously put a big disclaimer to let people know about the risks of each specific beta build.
Difficulty for customers to find my products
Currently, I sell my products on multiple platforms: SpigotMC, BuiltByBit, Polymart.
This leads to confusion because they can't easily peruse my entire product catalog.
Solution
I decided to create a simple website that provides a comprehensive list of all my products across all platforms in a single location.
New Spigot plugins projects
I'm starting to work on new ideas and publish them. The new Minecraft updates provide some interesting features I'm taking advantage of.
Some clear examples are RPGDrops and ChestPreview.
Other projects
Narcolepsy - horror videogame
This year, thanks to the new PC, I plan to finally release a working alpha version of my upcoming co-op horror game which I've been delaying for the last 7 years due to the lack of time and motivation.
Issues with the project
The main issues with this project are the fact that Unity 3D team is constantly messing up with the engine on each update, causing a very slow experience. With the new PC I hope to be able to work faster as I waste a ton of time building the project due to how slow the whole Unity 3D engine works.
Another solution is to completely discard Unity 3D and move to another engine or make my own engine, since the game uses PS1 styled graphics, so no fancy RTX Unreal Engine v666 is needed for this kind of projects.
Conclusion
Wishing everyone a great year in 2024 and thank you for being a part of my community.