Rewrite coming up
It''s time for a rewrite of the notes
Stuck in the Glue
For what feels like the longest time (it's only been some months) I've been working on rewriting the old ideas into Obsidian, thinking now they're finally going to be good notes and atomized, so we can start coding faster. But over time and changing the AI instructions multiple times, developing the powershell to help me move information between AI and Obsidian to avoid feeding too much information into the AI (which just eats tokens), I've once again hit a valley of glue. Not impossible to get through, but if I dont change direction, it will just take that much longer. The AI also either didn't get enough information or had to get the entire notes and subnotes to get the full understanding. But I noticed that the AI seems to have added a lot of fluff to the notes over time. While in the beginning, they were short and concise, now the notes it generates contains a lot of information about possible options for functionality and other well thought out data, but all data that is only relevant when actually coding, not for the think/rewrite work we are doing at this time.
Another Rewrite
So I mulled it over and decided that every single note will need to be rewritten once again, separating chaff from silver and gold. I need to actually break up the atomic notes into sections, so the powershell script know what parts of the note it should extract for AI comprehension, but leaving behind all the chaff, that has no real effect on AI understanding of the project. But chaff that is actually gold, once we start programming.
Dammit.. I've just spent the past many months rewriting and now I find out I need to do better, again. This is what I've been spending basically the past two years doing. Writing, then rewriting into a better format, then doing it again, and again... but every time the project changes. Every time we (AI and I) discuss, elaborate, develop, invent... and over the past two years the project has steadily and in leaps changed from simply a bunch of plugins to being a full logic-part gameplay engine administrator, but for plugins. Extending the basic game to be able to do things that the original game didn't even intend to do.
That is the nature of plugins. Making games do things they couldn't before. My framework just takes it to an entirely new level.
Choosing What to Rewrite
The remaining old ideas have been sorted into "rewrite now" and "wait till the new structure is implemented", because if I had to spend time rewriting all the rest, then it would take many weeks; and those weeks of work would require another rewrite into the new format.
I only need to rewrite 1600 lines of text into atomic notes, because those old ideas touch the core framework; before I can plan and reformat the existing notes. If I chose to rewrite the rest under the old format, I'd have to rewrite another 11.000 lines, twice. The rest are "only" plugin-level ideas, so not some that are likely to alter the core framework, but just ideas that are ment to be built as plugins using the framework.
Still Called ADK
Oh well... time to get started! But first I need to reset ChatGPT, because we just spent the past few hours discussing whether a different name for the framework would be a good idea. Unfortunately all the good options are either taken or so widely used, there'd be no real value in the change. I'm just not sure that "ADK Framework" is the best name, but it's what I've got at the moment.