Giter VIP home page Giter VIP logo

not-dnd-5thsrd's Introduction

Not DnD 5thSRD

The following is NOT a copy of the DnD 5th SRD.

It is a creative expression, of a common set of rules, for another realm. That is losely compatible with the rules of Dungeon and Dragons.

This is built using the collabrative effort, of human's and AI.

Where are the RULES?

Everything is in the docs folder here

For example, here is the rules on "Taking rest"

Not DnD 5thSRD

Taking Rest

Though they may be brave and courageous, adventurers cannot spend every hour of the day in the midst of exploration, social interaction, and combat. They must take time to rest, to sleep and eat, to tend to their wounds, to refresh their minds and spirits for spellcasting, and to prepare themselves for further adventures.

Extract from a major game

Resting

Heroic though they might be, adventurers can't spend every hour of the day in the thick of exploration, social interaction, and combat. They need rest--time to sleep and eat, tend their wounds, refresh their minds and spirits for spellcasting, and brace themselves for further adventure.

It maybe similar, but the wording is different.

this @LegalEagle video link is here for no reason in particular

LegalEagle-thumbnail

https://youtu.be/iZQJQYqhAgY?t=373

I like this video, and you should see it ๐Ÿ˜‰

How was this made?

I have gotten OpenAI GPT3.5 (aka chatGPT), which apperantly has been trained on the D&D rulebook. To write every page, in its own "flare".

This is done with either of the following techniques

  • Given inspiration materials, asked it to rewrite the rules in a different style and format
  • After finetuning with the various source materials, asked it to write the rules for individual spells, characters, classes, etc

The specific's of which page is done using which method, will not be disclosed. I will regenerate the pages if required, for the method that is more "legal friendly".

The process is entirely automated, as such as a human, I have not yet properly edit and proof read all 973 documents that were generated. ( I rushed this whole thing out overnight X_X )

PS: It took over 4 hours, for the AI to rewrite every page, one page at a time. Which I had to do about 3 times, due to silly coding bugs on my end.

Terms list

For the sake of avoiding confusion, with another well known game franchise.

The following are clarification, for common confusion, with potentially copyrighted or trademark terms.

Terms we use Terms, that some people may confused with Source Materials for our terms
Game Master Dungeon Master https://en.wikipedia.org/wiki/Gamemaster
Realms Planes https://en.wikipedia.org/wiki/Multiverse
Blindvision Blindsight https://en.wikipedia.org/wiki/Blindsight
Darksight Darkvision https://en.wikipedia.org/wiki/Night-vision_device
Truevision Truesight https://en.wikipedia.org/wiki/Eagle_eye
Baluran Gate Baldur's Gate https://en.wikipedia.org/wiki/Baluran_National_Park
Waterhelm Waterdeep https://lotr.fandom.com/wiki/Helm%27s_Deep
Sea Horror Aboleth https://en.wikipedia.org/wiki/Sea_monster

Note that the above is a limited list. The full (work in progress) list will be updated here

Is this legal, what is the license for this?

Honestly I do not know - I intend to put this under MIT license or Creative Commons - Attribution. But I do not know If I can.

This combines 2 giant legal furball.

  • copyright vs trademark See legal eagle full video
    • We would need to adapt all of this, to ensure it does not trip over any trademarked term
    • Our best approach, is to ensure all terms used, has origins from history (outside D&D, or use alternative terms)
  • Is an AI output copyrightable? This question is up in the air for the legal community!
    • I proposed a possible solution to treat AI models and output as derivative works of their previous steps here
    • Specifically in this case
      • if an AI model was used to transform copyright materials, what is the copyright of the output?
      • if an AI model was trained via fine tuning on copyright materials, and rewrites the output with some prompting, what is the copyright of the output?

You can see my article for way too much details on the AI issue : https://substack.tech-talk-cto.com/p/how-opensource-style-licensing-can

So until those are resolved, treat this as a dump of various WIP stuff.

If this is considered derivative work, it would be probably using the something like the OGL license If this is NOT derivative work, it would be MIT license or Creative Commons - Attribution.

Also to make it clear in advance, if anyone is willing to "take legal responsibility", of converting this to MIT licensing or Creative Commons - Attribution. I will not be persuing you, as long as you give attribution to this project (and its other contributors), as per the spirit of the MIT / Creative Commons - Attribution terms.

I am not a lawyer

So what can we do next to make this "Creative Commons - Attribution" compatible ?

  • Step 1) Review and remove any trademark / copyright infringing term.

    • The AI draws inspiration from too many sources, including copyright / trademark infringing sources.
    • The goal of this project, is to ensure all source materials, and terms, can be drawn on a common public domain heritage (eg. wizards is a noncopyrightable term, so is elves), that is free of copyright and trademark.
    • Rule of thumb, if you find the term used in 1800s literature (or earlier), we can use that as the source
    • Update the terms table accordingly (we might need to make this a seperate document)
    • To aid this process, the docs was scanned against an npm dictionary word-list, with unique words listed in terms/unique-words.txt
    • Also we must, have a human author edit every page, with atleast 20% changes.
      • This allows us to assert that atleast, the human authors changes are copyrightable under MIT licensing (until we have an answer regarding AI and copyright)
  • Step 2) Cleanup and fix any formatting issues in the various documents.

    • GPT had limits, as such I had to splt up chunks part by part, these may cause formatting issues within the same document
    • We should do a sentence to sentence check, that no sentence is copied directly from existing copyright materials.
  • Step 3) Make it builder-ble into a website / PDF format like : https://github.com/vitusventure/5thSRD

  • Step 4) Get the blessings of a lawyer? That we are good to go?

    • Hell, does this technically count as parody?

Once all these are resolve, maybe we would want to call it OpenSRD ?

Terms for pull requests

If you pass me any pull requests, you release all your copyright claims and licensing terms for your changes. I will keep your changes in git history, and ensure your changes are tracked as a contributor.

If the end result can be MIT / Creative Commons - Attribution licensed. You will be named as a contributor attributed under one of the above licenses (not decided which) If the end result is considered derivative work, it should be under OGL, and you have git history and credit (sorry)

More food for thought

You may want to consider the following, video - from a smaller D&D themed channel, regarding OGL

https://www.youtube.com/watch?v=8X8BS-2lP1s&ab_channel=ZoranTheBear


Like what I am doing?

I am Eugene from Uilicious.com and I go by the online handle of @picocreator

  • Consider signing up to my substack newsletter : https://substack.tech-talk-cto.com/
    • Random topics regarding AI, engineering, and startup.
  • Or support my small startup : https://uilicious.com
    • an easy to use low-code website testing tool, with AI !
    • useful, for constantly testing your website, for you !

not-dnd-5thsrd's People

Contributors

picocreator avatar

Watchers

 avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.