https://preview.redd.it/1ef6qgucwuih1.png?width=944&format=png&auto=webp&s=5fcaabb260ce8966f4b88bad26dbc3bc4d288f21

I wanted to follow up on a post I made a couple of months ago about why I started drinking a gallon of water a day.

The short version: a friend of mine has had an eGFR of around 17 for five years, and he told me he's avoided dialysis because he drinks about a gallon of water every day. My situation was different, I had a kidney removed due to cancer, and my remaining kidney (see the chart) had an eGFR that kept dropping. My blood pressure was around 170. I went through a lot of medications to get it down to about 150:

• chlorthalidone 25 MG tablet
• cloNIDine 0.1 MG tablet
• amLODIPine 10 MG tablet

But my eGFR was still falling. People telling me "your kidney should recover" wasn't helping when I was feeling all the negative effects (really tired, shortness of breath, ankle and leg swelling, really bad Anemia (I was always freezing).

My personal theory is that the extra water is flushing my kidney the way you'd flush a filter, so it can work better. I don't know if this makes any difference, but I drank all my water from those hydrogen water bottles they sell on Amazon (I know a lot of people think they are just a scam but I figured what the hell I'm still drinking the water...).

I had my nephrologist appointment today. He said I should be fine, and that my kidney may actually keep improving for up to a year. So apparently, I didn't have kidney disease after all, I just had a kidney that wasn't working.

I'm gonna keep drinking the water even though it is a total pain and I have to stop drinking by 5pm or I have to get up to go to the bathroom 7 times a night and it is hard to get much sleep.

Everyone here knows the pain. You're 40k words in, you ask the AI something about your story, and it has no idea that the character it's describing already died two chapters ago, or that her eye color changed somewhere along the way. The usual fix people reach for is RAG (chunking your text and doing a similarity search), but that only works for "find me a passage about X." It's useless for the questions writers actually care about, like "which characters show up in chapter 4 but never speak again" or "what did Alice look like back in Act II." Those answers don't live in any single paragraph. They live in the connections between things.

So for my open source app AIStoryBuilders I added a knowledge graph instead. Basically the story gets stored as a web of characters, locations, timelines, chapters, and paragraphs, all linked by their relationships. When the AI needs to know who appears in a scene or how a character is described in a given timeline, it just walks those links and gets an exact answer. No guessing, no "I think it was around chapter 5."

The fun part is continuity checking. You can ask it "is there a continuity problem with Alice between chapters 3 and 7" and it'll actually pull up the conflicting paragraphs and offer to fix them. And before it changes anything in your manuscript it shows you the edit first and waits for you to approve, so the AI never quietly rewrites your book behind your back.

The other thing that surprised me is how cheap this is to build. No fancy graph database, no extra server. It's literally a JSON file sitting in your story folder. The app is free and open source if anyone wants to poke at it.

I wrote up the full technical breakdown here if you're curious how it works under the hood: https://blazorhelpwebsite.com/ViewBlogPost/20086

https://preview.redd.it/e8v1guurc96h1.png?width=604&format=png&auto=webp&s=a85c0108c85ffeddd06fea31f71cd602999e7c3c

I updated my free ebook Blazor Succinctly and the new edition is out. It covers the core of Blazor and then builds a working Help Desk app so the concepts aren't just isolated snippets — auth/authorization, CRUD, forms and validation, email notifications, and AI features like smart paste.

It's completely free (online reader, PDF, and Kindle), and the full source is on GitHub.

Book: https://www.syncfusion.com/succinctly-free-ebooks/blazor-succinctly/introduction Code: https://github.com/SyncfusionSuccinctlyE-Books/Blazor-Succinctly

Happy to answer any questions about it.

Sorry for taking up space, but, after watching a ton of YouTube videos about "Don't use AI to write stories" I just had to write this...

Let's start with a calculator. You punch in 847 × 293 and get 248,171. Quick question: who did that math? You, or the calculator?

• Now picture a typewriter. You sit down, hammer out a letter, and pull the page out of the roller. Who wrote those words? You, or the typewriter?
• Open a Word document. You type out a report and the grammar checker quietly fixes your comma splices and flags a passive sentence. Who authored that document? You, or Microsoft?
• Now use ChatGPT to draft a short story. You feed it your idea, your characters, your direction, and out comes a draft. Who created that story? You, or the AI?

People are so consumed with detecting whether AI touched a piece of text that they've lost the plot.

If a human initiates the work, the result belongs to that human.

Full stop.

The whole debate about AI not being able to hold a copyright is ridiculous on its face. It's always the human who should hold the copyright, because the human is the one responsible.

Imagine sending a threatening email to the President and then telling the Secret Service, "Oh, my chatbot wrote that and sent it on its own." Let me know how that defense works out for you. I'll wait.

The reason I actually care about this is that the punchline of all this hand-wringing is, "Don't use AI." That's absurd. It's the same energy as telling early writers not to use typewriters, or telling accountants not to use calculators, or telling everyone in 1995 not to spell-check their documents because it's somehow cheating.

So did I use AI to write this blog post? Of course I did. I wrote down hundreds of words first — the ideas, the examples, the argument, the attitude. Then I had AI clean up my half-finished sentences and organize the whole thing into something coherent. Then I told it to cut the fat so the piece moved fast and didn't bore anyone.

I produced this. The AI just helped me type it.

Years ago I built a free open-source tool called AI Story Builders. I recently added the ability to chat with your story and also update your story through the chat interface. I was able to add this feature because I create a knowledge graph of the entire story. This allows the LLM to use function tool calling to see the entire story no matter how big it is. If interested in the technology behind it, I wrote a blog post on my technical site: More Powerful than AI RAG: Building Lightweight Knowledge Graphs

This is mostly about AI, but since all my samples are in Blazor I felt it was appropriate to post it here.

I work on mostly business applications and use a lot of retrieval augmented generation however I recently discovered the power of knowledge graphs and recently wrote a blog post about it and would love to hear feedback.

More Powerful than AI RAG: Building Lightweight Knowledge Graphs

Basically if you have any data schema that has related entities you can easily build a knowledge graph saved as simple in memory json, and expose that to your AI through function tool calling.

Then allow your end users to ask questions and query that data in ways unachievable without a knowledge graph. This can also be used to update and make changes to your source data.

Quick background, I've gone to college to learn to write and I have written a lot of fiction and nonfiction over decades.

I've also created AI writing programs. My main concentration now is on quality. I believe the reason why most AI writing sucks is it was trained on 'everything' and in general 'everything' is actually is pedestrian.

So, AI is more like a enhanced grammar checker it is incapable of producing what I call 'Quality' and by quality I mean writing that really grips you and moves you.

I am focused right now on the essence of quality so that I can feed these guidelines to the AI. I know there are prompts based on things like 'save the cat', but following those rules just produces 'decent' not gripping and emotionally moving on their own.

I am interested in thoughts on this.