Agreeing on standard transliteration files to use

[Koen G]

In another thread, Rene suggested that we might better use standard transliteration files. This sounds like a good idea to me.

However, I have no idea about the variation in quality between the various options. I am just using Takahashi out of habit, but will gladly switch if this appears too flawed.

What are peoples' thoughts on this?

[Aga Tentakulus]

I have personally worked with Stolfi. But it did not convince me.
But what I would be interested in is how it looks like when you compare the different transliteration files. How they behave among each other. Actually they should be precise.
Example:
You are not allowed to view links. Register or Login to view.

And comparing them with some languages. For example: Latin, Italian Spanish, French German. Maybe also English and Slavic.

Translated with You are not allowed to view links. Register or Login to view. (free version)

[bi3mw]

In my opinion, the files on Takeshi Takahashi's website are a good choice, but unfortunately he didn't edit some folios.  You would have to get them from the VIB ( with the missing capital letters ).

[Koen G]

In my experience, the difference between various VM sections can be much bigger than the difference between various transliterations. Though I agree it would be interesting to learn the impact of using one or another.

[-JKP-]

I don't think we can standardize transliterations until we know more about how to parse the shapes.

[davidjackson]

This is something that's popped up a lot over the years. Which is "best"? Who knows?
Through trial and error, transcribers have slowly come to realise that we can't map Voynich glyphs to a transcription alphabet, because we can't even agree on what a glyph is.
Some years ago I proposed merging different transcriptions and identifying problem vords which could then be manually sorted and fixed. Nothing ever came of it, mainly through my own inaction.
I still think it's a good way forwards. Mix them all together, using one standard transcription character set, and see where the different files don't agree.

[-JKP-]

I think there are two d shapes.

I think there are three m shapes, maybe four.

I think there are macrons. Which transliteration alphabet includes them?

I think l is misunderstood.

There are other matters, as well, ones that go beyond shape, but it takes more data to explain them and I can't do that during the workday.



Part of the problem is that many researchers (including most of those who have presented "solutions") have little or no experience with reading medieval text. Some are not even familiar with calligraphy. It takes a certain amount of experience to interpret the shapes within a medieval context.

It's too soon to standardize.

[-JKP-]

I'm not even sure standardization is necessary.

If you are feeding plaintext through a visualization app or other app, it doesn't matter whether you grab Plaintext A, B, or C as long as you identify to readers which one you used.


I doubt if there's any way to resolve a plaintext that treats gallows chars as single glyphs with one that splits out various components. It's probably necessary for them to co-exist until we know more about how to parse the glyphs.

[Anton]

I think there are two d shapes.

I think there are three m shapes, maybe four.

I think there are macrons. Which transliteration alphabet includes them?

There's no problem in addressing that by means of XML or JSON, although the description will cease to be human readable (remaining of course machine-readable).

But this requires new transcription effort of course.

[-JKP-]

I agree that it can be addressed. Computer software is very good at this. In fact I have my own transcription format that generates multiple options of plaintext.

But in terms of standardization, the way the current plaintext variants (Takahashi, etc.) are structured, it is less easy to accommodate these differences.