Bay 12 Games Forum

I really feel like it bears repeating - the fact that Simon Swerwer's dwarven lyrics are included in the official soundtrack of the steam release (and especially the one about the most famous DF story - Kogan Usan - Boat Murdered!!) lends a huge amount of credence and legitimacy to his interpretation of the dwarven language. Short of the Word of God stepping in himself to make it clear, I think this is the most robust endorsement of any dwarven interpretation we've ever seen, and is as close to official as we have ever had.

It's true the 'languages' of DF are underdeveloped and basically just giant gloses of English, but as the OP said Simon Swerwer used this crude base to form a fully-fledged language you can form sentences from - and indeed full song lyrics. With no inflexions, articles or prepositions to work with, it would be classified as an isolating language where relations between elements of a sentences are fully determined by a very rigid word order or can be inferred from context. Also, by necessity the differences between parts of speech (verbs, nouns, adjectives, adverbs and such) are very blurry and porous - an adjective can be used as an adverb, a noun as a verb, etc. and it all depends on context.

In that respect Swerwer-style dwarven has a very concise, implicit and terse quality that reminds one of Classical Chinese, which is an interesting development.

I also thought it reminded me of Chinese, and thought about bringing it up, but couldn't decide how to word it. Perhaps dwarven is pictographic, and their engravings are more literal than we've ever thought. Perhaps "dwarven language" as we see it written in game is merely a pinyin, or a latinized pidgin.

Haven't you guys seen this thread? It's not a pronunciation guide, but I think the pronunciation is obvious enough with the diacritic markings present. The lyrics sung by Simon Swerer seem to be close to the inspirations for dwarven languages, like Hebrew and other Cyrillic derived languages.

I remember attempting to train an LLM on that specific thread a while back to generate dwarven content, but ultimately failing. Probably because that's not how LLMs actually work, I now realize. I wonder if it would be possible to use AI now to help generate and understand the dwarven language.

Thanks for the thoughtful reply salmeuk, especially liked how you walked the fine line and managed not to criticize either of them, haha. I think you are mostly right and agree with you.

Look forward to your streaming. I'd play it myself, but as someone mentioned elsewhere, it kind of tends to crash, haha.

fill it with sand

this is harder than you might think, because an infinite amount of sand bags can be placed on each tile.

i'm sure you're up to the task

Nuclear fusion

You mentioned the color-matching of cave floors to their stone types being distracting, and that got me thinking: what's up with rock layer generation?
I know in earlier versions of DF (40d?) that the rock generation - or at least the metal generation - used to be different. From a geological standpoint, the generation today is quite strange, with giant ovals of different rock stacking over each other like the biggest layer cake on earth, from the circus to the surface. With every map tile, you basically know what you'll always get: a base square of layer stone, a giant oval of a different layer stone in the middle, and zero to several ores zigging through it.

Why is it done this way? Is this rock gen a holdover from the 23a days or such? Any plans to model, like, igneous intrusion, plate tectonics, stratified rocks, with the rock-tiles themselves in-game? Do geologists ever send you letters? 

cheers, thanks for the game toady. you "rock" too.

Quote from: Nihilich on July 02, 2023, 04:13:30 pm

2) What is the nature of the code-sharing deal you worked out with the DF hack team? Are you taking precautions against potential copyright issues?

I think the short answer here is that there's no code-sharing deal with the DFHack team.

For about a year now there's been a very small testing group that get an early release to bug test, and then what will be the public beta a couple days early. More recently, members of the DFHack team are in that testing group, so they can prepare for releases, and Tarn and Putnam are pretty accessible in that chat for specific issues that come up.

Nevertheless, they're almost entirely only getting a few days early access to public releases. I think it's underestimated in the community just how very good the core DFHack team is at putting things together from the same releases everyone else gets. Early access lets them release very close to the public release time.

Any access to actual code appears to be limited to occasional small snippets, but it having Putnam available (who's historically been active in DFHack) to explain this or that has probably helped a lot.

I'm not part of the DFHack team but I'm in testing, so it's possible I've messed up some small detail, but that's how it works in general.

Thanks for the reply, & i'm glad to hear that, though I was most worried about code flowing the other way - ever since that whole unfortunate meph thing ://

it's time!! it's here!!! the music update is here!!! we can (theoretically) use sound sense again!!!!!!!!!!!!!!!!!!!!