I just installed newLISP, and see that it can use OpenGL.
My idea is to make some program which can convert text into movie or game.
Of course, this is some Artificial intelligence-related,
so though I did my program in Visual basic, and it works more or less -
I was lately thinking how can I do anything related with something more complicated like part-of-speech tagging,
making graphics animation (and there is related Game-AI),
so I was told to use something like C++ as standard for programming,and also - Lua for graphics.
But what do you think - is it still better to learn lisp instead?
As far as I understand, there are two main languages - Lisp and newer Prolog,
and they are for AI.
So they can do natural language processing and some machine translation are done using lisp prolog,
and also there are some artificial intelligence models for game characters
and they must also use something like lisp and prolog (say in chess game).
But the thing that I do not know neither lisp nor prolog (though I tried to learn even Assembler, so lisp and prolog might be relatively easier).
I find that lisp is closer to natural language even in constructions, in code,
so that is also better, becasue I am somewhat tired of syntax of general languages like basic and c asm - doesn't matter -
they are very "machine" languages.
but people would say that lisp or prolog is also hard to learn, so I 'ld like to ask - is that true?
And especially if I compare that with learning C for graphics and text-processing (part of speech tagging etc)?
What I need first is some lisp or prolog source code which can read a text, say in English,
make part of speech tagging, and then give out some result, maybe even with relationships between words,
between multiple words (say, noun-phrases, verb phrases, adjective phrases, numerals phrases etc)
- this is done by machine translation software - they do this to make text in another language.
But for my program I do not need to translate into another language.
Rather I'ld need to "translate" into language of images or pictures and animations -
just like that is done not just in chinese or egyptian, and not only in pictorial dictionaries,
but in chess.
Chess programs can open some PGN files, and replay the recorded games
which have same "sentences":
- number of move
- figure to act with
So say if I have a source in lisp which can find nouns and verbs in sentences,
that can be converted into some "chess notation",
and thus noun+verb can become some openGL animation (say it is done using newLISP
or some other OpenGL supported lisp language/dialect.
I choose OpenGL as crossplatform, becaue some directx is just for windows,
so it is limited to waste time on this microsoft-only technology).
Some of related technologies or example programs or games are:
- widgit. search google for drupal widgit online. THere you type words, and images immediately appear
- can convert even abstarct words like prepositions, articles, numerals, adjectives, abstract nouns - into pictures.
- cannot do part-of speech tagging.
- only 2d graphics, though good - vector images
- not free, only demo available, so I like some opensource free alternative (hope to find 3d free objects to be converted say from 3ds max format into some opengl etc)
- biggest minus - no animation. so this is not game nor movie. I can get something like comix from text using widgit, but not animation.
type a word and get a animated character.
minuses - not free, cannot use whole text as input, 2d
another one, good:
spinnaker story machine (by the way, anone knows similar programs or/and games? please let me know if there are some!)
- can do kind of "part of speech tagging" for some 45 words like nouns, articles, prepositions, verbs, pronouns etc.
-supports animations. This is very nice. One can replay same story many times, objects appear in random place like in scribblenauts,
and animation happen. Simplest moves - go, walk, jump, hop. There are also some like "eat" etc. which are done automatically also if one is on distance -
subject goes to the thing to eat.
- dead thing can "eat" living. Dead things also go. THough it is not so bad - we know in mythology and religion - even sun can move, there are gods of different natural elements - like god of water, rain, god of fire, and we know that soul or life is there not only in plants, but even stones can grow etc...
SO, if in spinnaker "Apple" can go, or tree or flower can move - that is rather good idea. The thing is that spinnaker story machine doesn't analize there is noun animate or inanimate.
- it is 30 years old game. So graphics is too simple, though there is color version.
Thus - code is very small - just some 16 KiloBytes, and it could run on some 3MHz computer as I understand...
Well, anyway, I'ld like to combine those technologies, including chess PGN-files-anumation, widgit "text-to-images", story-machine-animation etc - into one.
THere one can write a text, and match it with some characters or actors who speak that.
But it done manually. It is no better than just some game creator where one has to do that manually.
Still they want money for that, and they call it "movie from text automatically", though it is not really so.
Next one is:
This is some programming language, but it has animation, 3d objects, and something closer to AI,
but of course syntax there is very tought - just like in most languages.
Use "." to separate noun-verbs etc. Why not use " " instead?
Some other similar programs (Scratch, including for second-life) are there, but these might be enough to understand the problem
So I'ld like to discuss all this.
Are there any people who who use lisp now - and writing similar program?
Or say prolog? Or maybe other programming languages, doesn't matter.
But I'ld like to know more about especially some open-source and/or free programs with simialr technologies.
I know that google has free translation from one language to another.
So such technologies to analize text into parts of speech etc - must be avalable in some lisp-prolog etc,
so where can I get that?
Or where do I find such communities or/and projects?
Maybe We can cooperate and do some more interesting project together (yes, text->3d animation)?