Thank you for the big write-up and your reasoning, I still (only slightly) disagree but I can also see that it's not a decision you made without much due consideration! * I also wanted to apologise if any of our nagging about recording got annoying at any point! I've absolutely loved this series so far in any case 💜 *(Edit: just finished the episode, I think this was a very good way to do the language decoding stuff, I still think showing notes for some of the holy cross puzzles would be good but this is more than satisfactory for this bit of the game)

I think this video was perfect and hopefully nobody is upset that they didn't see you studying and working on the writing system. Having gone through that process myself, I fully understand you not recording it and it would not have made for good videos. You did a great job explaining how y'all solved it, and then how it works. Also interesting to see your process and how it differed from mine! I got the same first words "a", "the", and "to" and guessed on "button" as well, but I also got "this" from the story page and focused on trying to read those pages rather than the tutorial pages. Figuring out "press and hold the button" was a great idea! Also, loved the playframe logo written in trunic at the beginning :)

The video is perfect! The extra editing work by Carrie is absolutely amazing, and this write-up makes complete and total sense. Honestly, the world doesn't deserve you two.

This makes total sense. I for one appreciate the effort you put in to make sharing the story of this fun and engaging for us. It makes me smile to imagine you and Carrie, chilling on the couch, pen and pad in hand, having a chill time while you recuperate. You guys are entirely too cute for your own good. Finally, congrats on the accomplishment. I may dip in and out, for nudges in the right direction, but I'm not sure I want to spoil this entirely for myself. Dan Floyd is good at games. Let none say otherwise.

Asking for that full seven hours of figuring it out remind me of when people wanted the full Castlevania Bestiary Unraveled.

Absolutely helped me to understand. Those visual cues then aided by audio cues go a long way. I could not have figured out that it's phonetic based language by myself.

it's so fun to write in this language once you get a good grasp of it~!

Where does it pop up in trunic? Huh?

@@Blubbpaule 2:08

@@rocovailo2862 My brain me tells it's the wireframe of a cube as viewed from a corner.

​@@techstuff9198you are correct. It is actually based on the isometric wire frame of a cube (or really mostly a rectangular prisim, but there are a few instances in the manual where the text is cube shaped, and understanding that is important because some of the glyphs look slightly different as a cube).

did you see the playframe logo at 2:09

@@linuxstreamer8910 yes! That was such a lovely touch

@@linuxstreamer8910 hahaha I had not - thanks for pointing it out! Obviously easier now with front files available, but further shows the level of polish these two do 💚

Based on how the episode went, that overlay was mostly Carrie's doing, so lets all say "Thankyou Carrie!"

Bravo

may of the fox ghosts are actually there to help teach the language as well. One of the characters does the same thing. The one lying on the floor. Zzzzzzzzzz

Another fun one is the fire bomb description. The second line of it is just "ow ow ow ow ow" lol

My favorite one is the guy by the well going “well, well, well…” I never learned the language during my play through, but that never stopped me from knowing what he was saying.

Let's move on from writing systems in general to Tunic itself in particular, and start by looking at English phonology. The moment you recognize that it's not a cipher for Latin script (what we normally write English in), you probably should have shifted to using IPA immediately [3], because English spelling is pretty damn atrocious thanks to the Great Vowel Shift [4]. If you're not familiar with phonology in general, Wikipedia's page on English phonology is a pretty good introduction for figuring out how you might transcribe English speech. English has 24 consonants, written in IPA [5] as /m/, /n/, /ŋ/ (the 'n' as in 'sing'), /p/, /b/, /t/, /d/, /k/, /g/, /tʃ/ ('ch'), /dʒ/ ('dg' like 'fudge'), /f/, /v/, /θ/ ('th' as in 'thin'), /ð/ ('th' as in 'this'), /s/, /z/, /ʃ/ ('sh'), /ʒ/ ('s' in 'vision'), /h/, /l/, /r/, /j/ ('y'), /w/. This list isn't arbitrary, actually. Consonants have a structure to them: there's the kind of action you're doing (forcing air through noise, briefly stopping the airflow, hissing around your tongue, or... it's complicated); where your tongue is (ordered here from lips to the back of your mouth); and whether or not you're vibrating your vocal cords as you do it (known as voice) [6]. I've sorted them by these characteristics. If you're designing a writing system from scratch, it would make some sense to make consonants that have some relation to each other display some sort of relation in written script. Does Tunic do that? Well, most consonants have a voiced-voiceless pair: /p/ versus /b/, or /θ/ versus /ð/, for example. In Tunic, each of these pairs turn out to be mirror images of each other, so /p/ is upside-down /b/, /f/ upside-down /v/, etc. Another feature is that the /tʃ/ and /dʒ/ look rather like it should be /t/ + /ʃ/ or /d/ + /ʒ/ (phonetically, you're shifting from one to the other sound in smooth motion). If you look at the script... it's a missed opportunity. The character /ʃ/ looks tantalizingly close to /tʃ/ + /s/, but it just doesn't work out. There's also a correlation between the stops and fricatives in the same place of articulation: /f/ is /p/ with a little line to the lower left that is the same addition you use to get /s/ from /t/. There's not enough pairs in English to tell me if this alteration was intentional or not. I don't see any other patterns. I will also note that the manual's "alphabet" at the end uses an ordering very nearly the ordering I used, although /tʃ/ and /dʒ/ are clearly in the wrong order [7]. On to vowels. Vowels in English are... complicated, especially because even the major dialects merge them in inconsistent ways, and Latin is poorly equipped to handle them. At a basic level, English has 5 main vowel sounds in "short" and "long" form. The short vowel sounds (drop them in "b*t"), in IPA [8], are /æ/, /ɛ/, /ɪ/, /ɒ/, /ʌ/, and the long vowels (use "b*te") are /eɪ/, /i/, /ai/, /oʊ/, /ju/ (the 'y' sound creeps in there in long vowel form). The short vowels are all monophthongs (your mouth stays the same shape), while most of the long vowels are diphthongs (you shift from one vowel position to the other). In addition to these 10 basic vowels, we have an additional basic monophthong (/ʊ/--like "foot"), an optional additional monophthong between a and o (/ɔ/ as in "caught" if you *don't* have the cot-caught merger), /ə/ (schwa, I'll get back here in a second) and two more diphthongs (/ɔɪ/ like "boil" and /aʊ/ like "how"). There's also several "r-colored vowels", as several dialects don't pronounce an /r/ sound here anymore (known as a non-rhotic accent), which I usually don't count as separate vowels but are distinct in Tunic's script. If you're going to make it distinct, you need 6 different /Vr/ sounds, as in "far", "fair", "fear", "fir", "fore", "fury" (sorry, "fure" isn't a word) are distinct sounds [9]. If you tot everything up, you should need 20 or 21 vowel sounds in Tunic, and realistically, you'd want to do 21 to preserve cot-caught distinction for those dialects where it matters. Look back at our table, however, and there's only 18. The missing sounds are /ʊr/ (as in "cure") and /ɔ/, while /ə/ and /ʌ/ have been merged. I have a pretty big issue with how Tunic phonetically transcribes English, and it centers on the /ə/~/ʌ/ merger. If you look at a chart of where vowels are pronounced, /ə/ is the vowel that sits in the dead center. Consequently, it's kind of a neutral vowel. In English, vowels in unstressed syllables tend to end up neutralized and shift into an /ə/ from their original pronunciation (this is called "schwa reduction"). There are some words where certain syllables are never stressed, even when the word is, and the vowel sounds in those positions are routinely identified as being purely /ə/. The canonical example is "about"--pronounced /ə.baʊt/). It's not really possible to stress a syllable with /ə/ in English. When you have words that are usually pronounced unstressed--"a" and "the" are good examples--when you do stress them, you usually change the pronunciation entirely to match. "a" is /eɪ/ when stressed and /ə/ unstressed, while "the" is usually /ði/ when stressed and /ðə/ unstressed. (The unstressed/stressed distinction was noticeable when you were going through the description of "a" being written as /ə/). I can see why the developers would prefer to transcribe words that are usually unstressed in their unstressed pronunciation, though it does feel weird to me. The problem is when you go from the word "a" to "button". "Button" is an interesting word because of its second syllable, which consists of, um, /n/. By itself. Kind of. Sometimes, there's a tendency to give the second syllable a little bit of a vowel, which is of course would be the standard unstressed vowel. So you'd expect to see "button" transcribed as /bʌt.n/ or /bʌt.ən/. Tunic transcribed it as... /bətɪn/, which is quite different instead [10]. My guess is that /ə/ and /ʌ/ were merged because most people would tend to transcribe those sounds as both "UH", but given that the game preserved the /θ/ and /ð/ distinction (both would be transcribed as "TH" and I can think of only one minimal pair where the distinction actually matters--"thy" versus "thigh"), and that it also distinguishes /ŋ/ from /n/ where /ŋ/ only appears where /n/ would appear before /g/, why it didn't go to the same length to preserve /ə/ and /ʌ/. [3] English does seem to be unusually averse to using IPA. In the dictionaries you read in other languages, pronunciation is almost always given via IPA symbols, whereas English dictionaries seem to tend to prefer ad-hoc phonetic spelling instead. [4] To make a long story short, we write English according to how it was pronounced roughly 500 years ago. Then almost all of the vowels ended up shifting pronunciation so that it makes very little sense, but the spelling of written words didn't change. [5] To clarify, anything I have in brackets like this: /p/ is meant to be IPA rather than English text. [6] There are other things you can do, but they don't distinguish between different phonemes in English, so I'm not getting into them here. [7] If you need help seeing it, the pattern is that voiceless comes immediately before voiced in every pair, *except* for /tʃ/ and /dʒ/. [8] I'm using General American as the basis for IPA transcription here. [9] Strictly speaking, for General American, you need an extra vowel sound that only crops up before the /r/ to distinguish these sounds. [10] It is worth pointing out that unstressed vowels can reduce to a sound somewhere in the /i/ or /ɪ/ region as well too, so it's not entirely crazy. But it's worth noting that dictionary convention for syllabic consonants is to use ə if a vowel is prescribed, and I would expect a different vowel to be pretty strongly pronounced.

yes i love that! makes it easy to remember.. the vowels always trip me up

it helped me to remember certain sounds to glyph thanks to some of them kinda-sorta resembling the english letter. Maybe not intentionally. But like the Kkk sound kinda looks like a K and the S sound swivles a bit to kinda seem like an S. That's how my brain helped remember it haha. Rrr was also kinda similar to the r shape, lll by far the easiest as it's just a straigh line, like a lower case L is.

@@Phoeniiix A trick with the vowels: The symbols for 〈eer, ere, ore〉 are just the symbols of 〈ee, eh, oh〉 with the bottom left segment removed

When I realized that the mirroring was a pattern, it helped me figure out the rest so much easier. I will admit I had a bit of spoilage on the fact that it was phonetic and which parts were vowels/consonants, but once I found that out, I determined to sort out all the sounds on my own with no further hints. And I did. :3

Indeed, the dev is from Halifax and sometimes it shows in the way Trunic is written. Thankfully it's only really an issue when trying to figure out what each symbol means, because you're often trying to compare two words which may sound the same to you and trying to find the symbol they have in common, but they don't have any because they don't sound the same to him. But when you're just trying to read you can usually skip a vowel that sounds weird to you (a vs eh for instance) and it still makes sense.

As a linguist (and ESL) this tripped me up a bit at first but once I realized there was a slight accent to the script it made i so much cuter. But also a bit harder to decipher, of course.

... can anybody give me even ONE example in modern media that I can google down of "DREAM" being said with a J in it?

@@ICountFrom0 it ends up sounding really similar in context, so it’s barely noticeable. (At risk of looking silly in public) try whispering the word to yourself with the different sounds instead of fully speaking them. Whispering makes your brain really look for context clues and you’ll see how weirdly similar the two phonemes can be in a similar context

sorry, but I just can't find anything that you can swap with a J, in the word dream, that sounds even remotely like english. If there's a particularly movie, or radio show episode, or famous youtuber who has that accessnt..

All of the symbols are in the book, but those last two are pretty rare.

@@syirogane Yup. I know exactly where they are in book too, as I had to help someone in finding one of those characters just a couple of days ago.