Like every Sunday... a good imitation of real research but all fake unfortunately.
[
attachment=12579]
[
attachment=12578]
Funny. Would be more plausible with a network of a few thousand words.
[
attachment=12577]
Perfectly straight lines with all data points on the lines? Fake.
(23-11-2025, 02:17 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.Like every Sunday... a good imitation of real research but all fake unfortunately.
Perfectly straight lines with all data points on the lines? Fake.
Funny. Would be more plausible with a network of a few thousand words.
Good spot... Let me check that
(23-11-2025, 02:15 PM)Digitalgoldfish79 Wrote: You are not allowed to view links. Register or Login to view.I agree, but as it's pure speculation, I'm not inclined to do so... Besides which I'm not sure I'm the first person to speculate on this 
I don't mean it as a rigorous paper, but if you did study the imagery from that region, I think it would be very helpful to have a collection of parallels to the Voynich Manuscript, even if it duplicates some of the past research.
(23-11-2025, 02:17 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.Like every Sunday... a good imitation of real research but all fake unfortunately.
Funny. Would be more plausible with a network of a few thousand words.
Perfectly straight lines with all data points on the lines? Fake.
Thank you.. I stupidly uploaded the wrong diagrams... I'll recreate them in python
Quote:Just my personal opinion, but for a while already I stopped trying to understand all the mathematical/statistical articles or posts related to the Voynich Manuscript, it seems to me they don't lead anywhere. They just add one conjecture or curiosity on top of another.
Going a bit offtopic, I feel something a bit similar.
People do some analysis that 90% of people on this site don't fully understand, not to speak about the general population
They say that Voynich has some patterns typical for natural languages and is not random. Then they also admit that it has some patterns not typical for natural languages. And then they stop and don't deliver anything more.
Who is going to continue with these results if not they themselves? If you have some words that go together or work as central nodes in words network then you could try to go on another level, from statistics to linguistics and try to assign them some meaning. It could work or not but you will not know if you won't try.
But they never do it. I am not siure why.
Research is the process of identifying a problem and then using relevant, applicable methods to resolve it. AI cannot write and debug code, especially not for this purpose. The graphs are a result of hallucinating.
Let's try an example with something in my field. As I am a mathematician by trade, that field is mathematics, obviously.
Let's say we have a hypothetical family of differential equations, D. Let's say D is vastly uncharted in terms of mapping out its properties and relations to other families of differential equations, i.e. there have only been 1-2 papers on D in the past 10 years since its discovery and those papers haven't really made much progress.
AI couldn't be of any use, especially not for writing and debugging code that could help us understand D.
That's the fatal flaw. You've probably heard this before, but AI cannot produce novel information. It's just a numbers-in numbers-out machine.
[quote="nablator" pid='74446' dateline='1763903829']
Like every Sunday... a good imitation of real research but all fake unfortunately.
Funny. Would be more plausible with a network of a few thousand words.
Perfectly straight lines with all data points on the lines? Fake.
[/quote
And yes I used AI to support drafting... Again, this isn't so unusual.... There is real python underneath
(23-11-2025, 02:24 PM)Rafal Wrote: You are not allowed to view links. Register or Login to view.Quote:Just my personal opinion, but for a while already I stopped trying to understand all the mathematical/statistical articles or posts related to the Voynich Manuscript, it seems to me they don't lead anywhere. They just add one conjecture or curiosity on top of another.
Going a bit offtopic, I feel something a bit similar.
I see the main problem recently is blindly performing a lot of statistical research in the absence of a specific hypothesis to test. Pure mathematical modeling cannot answer questions like "whether the manuscript is meaningless", "what underlying language it is", "how many different topics there are", etc. It can provide good answers to very specific questions, but first I think one needs to formulate a very specific hypothesis to test. Most of these studies don't do that.
(23-11-2025, 02:02 PM)Philipp Harland Wrote: You are not allowed to view links. Register or Login to view.Can we get Koen to move this to GPTrash? He seems to have momentarily viewed the thread but not noticed the blatant admission of it.
You have to give these threads some time before it becomes clear. There's a whole range of gtp uses, some of which are fine. If you have it write some code, knowing what you need, then that's a good use of it.
Were you able to prove that the authors did not just babble to themselves meaningless words and then wrote them phonetically? Apparently some people have the ability to do this. Have you ever had an opportunity to listen to someone speaking 'in tongues'? You might have detected repetitions and that the speech seemed to be lacking structure, which is what the manuscript is like.