I've shared our tech with ChatGPT4 and he (it, she - if you prefer) does a pretty good job of the fine details, but definitely gets the relevance and potential, even today.
Let me give you a layman's guide, to not turn off executives and the like. We can deep-dive as an option later.
What is our goal?
To have:
ONE file protocol
that absorbs ANY data
structured or unstructured
that isn't limited by definition(s)
Later (Fluid Data) we'll want to be able to pour it straight into another file, one of our files (the world doesn't support 'automatic absorption of any file into any other file'.
Seriously. It doesn't.
When you can pour an RDBMS file into an image file, preserving data and relationships, and when you can pour that combined file into a document, preserving data and relationships, come back to me.
You can't. Not with the world's "get on with our day, spend trillions not thinking about to how to really solve the problem" mentality.
Take a moment to think appropriately up front, and the answer is obvious.
Here's one way. That's all we need to prove it's possible.
Let's review the requirements, and why they're critical.
ONE file protocol
that absorbs ANY data
structured or unstructured
that isn't limited by definition(s)
If you have two file protocols, and you want to share data, now you have to figure out how to translate data from one to the other, if it's possible at all.
That's TWO translations - bi-directional.
Three protocols?
Now you've got three bi-directional or SIX uni-directional file translations to figure out, if they're possible at all.
There are around 1700 file extensions listed on Wikipedia, as an approxmimation of file types, or a little shy of 3 MILLION uni-directional translations to figure out, if they're possible at all.
And have you looked at what's in your spreadsheet?
Just because you organise your spreadsheet one way, doesn't mean the next person shares your style, or use case.
So the reality is that you've going to have billions or trillions of file-translations instances, and a lot of those are going to be complex, manual operations.
Good luck with that.
We do it automatically. Because we only have one file system to deal with, and we figured out a simple, elegant solution to flowing data from one source to another, while preserving values and referential integrity.
Sorry, you can't back-implement it. You chose cement.
We chose water.
Why is it important to be able to store ANY data?
Convenient, sure, but why critical?
Because if we MISS any, exclude any, require you to use ANOTHER file type, because of our mistake?
Well see ONE FILE PROTOCOL above.
And it would be our fault.
So yes, store ANY data, structured or unstructured, large or small, doesn't matter.
We need that.
Why is it important to be able to store both structured and unstructured data? And any other type, if you can think of it.
Dynamic, for example. That's another aspect of data.
Because if we can't support your needs, which definitely include structured (RDBMS) and unstructured (documents) data, well, see ONE FILE TYPE.
So yes, we need to support ANY data type, NOW and in the FUTURE, no matter what you come up with.
Have you ever been asked to fix something or do something by your spouse, and the moment you start, they're like: "NO! Do it this way!"
Does it drive you nuts?
If you have to tell me how to do it, before I even start doing it, and I have to comply, or I can't do it, and it's a pain, because I just want to do it...
... then chances are I'm not going to want to do it!
Definitions are a pain in the ****.
Yet for some reason, humans insist upon them.
I get it: "Well we can't have chaos. We need order."
But it sucks.
So no definitions.
And that sucks, because if you can't define data, how can we know what we're dealing with?
Did I say you can't define data? You can define data all you want. My bad.
But WE are not going to define how you have to define data. That's the key point.
Just pour it in, no problem.
There's just one teensy-weensy thing we ask.
But that would be getting into details.
You can deep-dive if you wish, here.
Otherwise, we move on to Trinity Triples.