Racket News - Issue 18

[pma...@linki.tools]

Racket News Issue 18 is here!
https://racket-news.com/2019/10/racket-news-issue-18.html

Grab a coffee and enjoy!
--
Paulo

[Hendrik Boom]

> --
> You received this message because you are subscribed to the Google Groups "Racket Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to racket-users...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/d4d-5dadf100-59-30cea940%40133288278.

Unfortunate that the Racket game jam got scheduled in the same month as NaNoWriMo (write a
novel in November) and NaNoGenMo (write a computer program that writes a novel in
November).

I'm involved in two of them already, and now there's a third one? Help!

-- hendrik

[Stephen De Gabrielle]

A NaNoGenMo submission counts as a game jam entry.

HTH

Stephen

To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/20191021190422.6fknnlvarvjngfy5%40topoi.pooq.com.

--

----

[Hendrik Boom]

On Mon, Oct 21, 2019 at 11:13:34PM +0100, Stephen De Gabrielle wrote:
> A NaNoGenMo submission counts as a game jam entry.

Even if it's not a game?

> To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/CAGHj7-LSgscrYafD9geZpuW2ZU8tChSxBEjkzWJzvpfJtNqdgw%40mail.gmail.com.

[Stephen De Gabrielle]

On Mon, 21 Oct 2019 at 23:18, Hendrik Boom <hen...@topoi.pooq.com> wrote:

On Mon, Oct 21, 2019 at 11:13:34PM +0100, Stephen De Gabrielle wrote:
> A NaNoGenMo submission counts as a game jam entry.

Even if it's not a game?

Algorithmically generated text is an important part of many games. 

There is even a language for procedural text generation: https://github.com/galaxykate/tracery - I kinda want to make a racket/RacketScript version, but I don’t have time at the moment. 

The game jam entry would be the code, not the procedurally generated novel. 

S.

To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/20191021221834.4oecz6suney6omam%40topoi.pooq.com.

--

----

[Stephen De Gabrielle]

I’ve updated the game jam description to add:

• Full games, game components or tools are all acceptable entries. e.g. tile editor, game map editor, procedural content generator, etc.

--

----

[Hendrik Boom]

On Tue, Oct 22, 2019 at 01:59:25AM +0100, Stephen De Gabrielle wrote:
> On Mon, 21 Oct 2019 at 23:18, Hendrik Boom <hen...@topoi.pooq.com> wrote:
>
> > On Mon, Oct 21, 2019 at 11:13:34PM +0100, Stephen De Gabrielle wrote:
> > > A NaNoGenMo submission counts as a game jam entry.
> >
> > Even if it's not a game?
>
>
> Algorithmically generated text is an important part of many games.
>
> There is even a language for procedural text generation:
> https://github.com/galaxykate/tracery - I kinda want to make a
> racket/RacketScript version, but I don’t have time at the moment.

Tracery looks like a tool to generate random sentences, with the option
to add new grammar rules as a side effect of using particular grammar
rules.

It doesn't look great for use in a game, just in automatically generated
nonsense, which is still what nanogenmo usually does. Sometimes the
nanogenmo-generated text fraws the reader in for as much as a page
before ie realises that this is nonsense. (yes, and a whole page is
quite an accomplishment).

What's desired in a too for a game is to take an abstract syntax of what
is to be said, and to do the necessary transformations to make it into a
grammatical sentence.

This would likely be a worthy task, even if it implemented a usable
but restricted language. THings like proper plurals in the right
places. The game designers would have to provide the conjugations and
such for any words not already in the lexicon.

I'd likely be working on a plot generator, which is a limit deduction
engine.

What I got last year was a program that with the right axioms, would
generate
Bill kills John
from facts like
Bill loves Mary
John loves Mary
suitably formalized:I had statements like (kills Bill John) and
(loves Bill Mary)

Unfortunately, from the same facts, it also generated
Bill kills Bill.
:-)

I've reformulated some of this in Racket recently. It turns out to be
surprisingly easy by defining macros that generate uses of 'match'.
And it's easy to squeeze in ordinary Racket code into the rules.
I'm going to have to make some more macros for purpose-directed
reasoning.

If a player is given a set of actions to choose from, which change
the state of the universe, the game could apply deduction rules to
find a response. The response would generate a new state and thus
new actions for the player.

-- hendrik

> > https://groups.google.com/d/msgid/racket-users/20191021221834.4oecz6suney6omam%40topoi.pooq.com

> > .
> >
> --
> ----
>
> --
> You received this message because you are subscribed to the Google Groups "Racket Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to racket-users...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/CAGHj7-LQdvYNtq%3DQsttnrhh4DgmiKmR1OfWH%3DuBh4HOQ0nbxOg%40mail.gmail.com.

[bruno cuconato]

On Mon, 22 Oct 2019 at 19:28, Hendrik Boom <hen...@topoi.pooq.com> wrote:

> What's desired in a too for a game is to take an abstract syntax of what
> is to be said, and to do the necessary transformations to make it into a
> grammatical sentence.

for the abstract syntax to concrete representation part,

GF(http://www.grammaticalframework.org/) might come in handy!

or maybe one would like to reimplement something like it in Racket...

[Hendrik Boom]

Looks interesting. But it may be overkill for text games.

-- hendrik