Request for roguelike thesaurus
Andrew Doull
There's a parallel thread discussing adjective lists for monsters at
the moment, which has prompted me to suggest a project that the
denizens of rgrd could contribute to.
The problem: roguelikes need lots of lists of words - a thesaurus if
you will - of common fantasy materials, in order to populate the
worlds with believeable yet randomly generated items. These lists
could be used to create place names and people names, much in the way
Dwarf Fortress does, as well as different item types (birch staffs,
bamboo staffs etc).
And unfortunately, nothing of the sort exists that is not encumbered
by intellectual property rights (Question: Is Roget's in the public
domain?). You can see this by searching the common open source
thesauri for the term obsidian and see that no results are returned.
Obviously, the roguelike thesaurus doesn't just have to include
fantasy tropes, but this is probably the best place to start.
Ideally, these lists need to be as unencumbered as possible: ideally
in the public domain (which allows immediate fork into GPL and BSD
style licenses). They would greatly assist anyone who started out
writing a roguelike, and that's what they are all for.
Places to start:
* Angband and variants have a large list of material types released
under the Moria / Angband license and the GPL. Check out the
flavours.txt list.
* Dwarf Fortress has various lists. I'll let someone contact Tarn
about their use and/or permissions restrictions.
* Any other suggetions?
Decisions to make:
* How should this information be structured? I suggest a thesaurus as
probably the best approach.
* Where should this resource be hosted?
* Is it worth trying to do at all?
Regards,
Andrew
Andrew Doull
Sorry to self-reply: but apparently it is: http://www.gutenberg.org/ebooks/22
It still fails the obsidian test however...
Andrew
Timofei Shatrov
tried to confuse everyone with this message:
>Hi,
>
>There's a parallel thread discussing adjective lists for monsters at
>the moment, which has prompted me to suggest a project that the
>denizens of rgrd could contribute to.
>
>The problem: roguelikes need lots of lists of words - a thesaurus if
>you will - of common fantasy materials, in order to populate the
>worlds with believeable yet randomly generated items. These lists
>could be used to create place names and people names, much in the way
>Dwarf Fortress does, as well as different item types (birch staffs,
>bamboo staffs etc).
Yeah, like we need more generic common fantasy roguelikes...
It's not important if your game has only 10 materials and 20 monsters. What's
important is that they fit the setting well.
--
|Don't believe this - you're not worthless ,gr---------.ru
|It's us against millions and we can't take them all... | ue il |
|But we can take them on! | @ma |
| (A Wilhelm Scream - The Rip) |______________|
Ray Dillinger
> On Sun, 31 Aug 2008 15:37:07 -0700 (PDT), Andrew Doull
> <andre...@gmail.com> tried to confuse everyone with this message:
>>The problem: roguelikes need lots of lists of words - a thesaurus if
>>you will - of common fantasy materials, in order to populate the
>>worlds with believeable yet randomly generated items. These lists
>>could be used to create place names and people names, much in the way
>>Dwarf Fortress does, as well as different item types (birch staffs,
>>bamboo staffs etc).
> Yeah, like we need more generic common fantasy roguelikes...
> It's not important if your game has only 10 materials and 20 monsters.
> What's important is that they fit the setting well.
My problem with a list like the proposed master list of adjectives
is that it could inadvertently work to destroy rather than build
settings and worlds; when you take things from all over, things that
don't belong together, the player has no sense of orientation or
place. At the very least I'd want to tag the descriptives with
codes for region and era so as to give game designers who care
about such things the option of having a setting for players to
enjoy.
If I'm playing a character with European accoutrements like full
plate mail, a shortsword, etc, I would find it very jarring to run
across a bamboo staff. Bamboo doesn't belong anywhere near that
part of the world, it's not a European plant. On the other hand
if I'm fighting with a kukri and a sharktooth sword, have a canteen
made of a coconut shell, and my base of operations is a lipalipa,
then the bamboo staff makes perfect sense but I'd be annoyed at the
game designer to find, say, a sack made of woven cotton or linen
cloth because cotton and flax are American and inland south Asian
plants respectively and don't belong in *that* part of the world.
Bear
Numeron
words which are similar (ie related to the same area or material) in
order to help expand item/monster sets, not in making a game more
generic. For example, there have been times where I have jumped onto
the net for villain name generators to help me expand some of my
unique monsters. They rarely fit my game, but I never actually use the
names they spit out anyway: the reason I do it is because some of the
combinations can definately give good ideas with mixing up words I'd
never have otherwise thought of (like mindstriker, which is now a
standard caster monster in lower levels of my RL)
If a developer decides to put a bamboo staff into futuristic scifi
roguelike, then while there might be some twisted abstract path
through the thesaurus that is actually logical: its the developer you
should blame for the idiocy of putting it in, not the thesaurus.
-Numeron
konijn_
> Timofei Shatrov wrote:
> > On Sun, 31 Aug 2008 15:37:07 -0700 (PDT), Andrew Doull
> >>The problem: roguelikes need lots of lists of words - a thesaurus if
> >>you will - of common fantasy materials, in order to populate the
> >>worlds with believeable yet randomly generated items. These lists
> >>could be used to create place names and people names, much in the way
> >>Dwarf Fortress does, as well as different item types (birch staffs,
> >>bamboo staffs etc).
> > Yeah, like we need more generic common fantasy roguelikes...
> > It's not important if your game has only 10 materials and 20 monsters.
> > What's important is that they fit the setting well.
>
> My problem with a list like the proposed master list of adjectives
> is that it could inadvertently work to destroy rather than build
> settings and worlds; when you take things from all over, things that
> don't belong together, the player has no sense of orientation or
> place. At the very least I'd want to tag the descriptives with
> codes for region and era so as to give game designers who care
> about such things the option of having a setting for players to
> enjoy.
Yes, ideally ( a pipe dream ? ) it should work like the visual
thesaurus. It should group the word lists/synonyms in groups ( science
fiction, middle age, eastern, 70's literature etc. ) If you go in
visual thesaurus for the word "supernatural" you'll see what I mean.
I only know one radial layout library that is open source and it's
java, so you'd need an applet ( like visual thesaurus ).
A thesaurus would save a lot of time for me and maybe others. I
replaced Mystic with Orphic in Hellband because it sounded too Eastern
in my ears and because Orphic is closer to the roman/greek origins of
Dante's Inferno.
Cheers,
T.
MrBudgens
>
> The problem: roguelikes need lots of lists of words - a thesaurus if
> you will - of common fantasy materials, in order to populate the
> worlds with believeable yet randomly generated items. These lists
> could be used to create place names and people names, much in the way
> Dwarf Fortress does, as well as different item types (birch staffs,
> bamboo staffs etc).
>
With materials for making wands in mind I fed in Obsidian, Birch,
Uranium, Steel, Glass. The resulting list is reasonably promising,
even if "upholstered" doesn't quite make the grade!
Not everything works well, but it's a good start.
Michal Bielinski
> http://labs.google.com/sets
>
> With materials for making wands in mind I fed in Obsidian, Birch,
> Uranium, Steel, Glass. The resulting list is reasonably promising,
> even if "upholstered" doesn't quite make the grade!
>
> Not everything works well, but it's a good start.
were postcardware, dragon and si. Impressive.
--
Michal Bielinski