Thank you very much for clearing that up! I like to have my Greek ↔ Latin transliterations finetuned.
(As to what purpose, I’ll leave that to the imagination of the reader…)
1 Like
Usually, I would recommend using 6L38 for reasons (although it’s very old), but you should be OK with 10.1. There are some thing that are a pain to port from 6L68, some things that need to be refactored, and some things that don’t work anymore (or I didn’t find the documentation to make it work).
But that things might very well not apply to you, and you’re starting from scratch, so… Stick with 10.1, I guess.
And also, the next version of Inform will add features for things that mainly require hacks or workarounds right now (notably IE-0016), so I prefer to wait a little before fully updating the French translation.
2 Likes
The two biggest changes from Ancient to Modern pronunciation are:
- β δ γ φ θ χ changed from stops to fricatives—but of those, only β and φ have a letter conveniently equivalent to their new pronunciation in English (or Latin), so the rest are still transcribed the same way as before
- The incredible number of vowels got simplified down to only five, with most of the Ancient vowels all turning into /i/ (the vowel in “seed”)
One of the famous pieces of evidence for these changes (both of them!) is that ancient authors describe sheep going “βῆ βῆ”, and beh beh sounds much more like a sheep than vee vee does.
9 Likes
What a wonderful piece of linguistic archaeology!
1 Like
I remember reading a thesis about the first documented greek language version of inform 6. Not sure how helpful that would be, especially considering that’s fairly under the hood but if you’re interested i can probably find the actual paper.
Γεια σου Γιώργο! @GJMen
Probably you are talking about this thesis from Sofia Delikostidi. I also asked about that above, and it looks like that I could use at least some part of this thesis for the translation of Inform 7.
However, the problem is that it looks like that the best option is to continue with a transliteration instead of using Greek characters, so I will at least need to modify it to work with Latin characters
At least in the Understand lines, yeah. Making it so Understand lines can include non-ZSCII characters is going to be a significant pain and will involve modifications all the way down to the compiler level.
They’re modifications worth making, don’t get me wrong, but it’ll be even more of a challenge than the already-monumental undertaking of translating I7 to Greek in the first place.
1 Like
For anyone that may have not seen that, @zarf made a change that solves the issue, and instead of 8-bit character arrays, 32-bit are used instead!
4 Likes