PROJECT REJECTED: unbgames

[Jeff]

A resubmitted project worth another look...

[rommelnc]

Hi everyone!

Who does not want to play smarter games? Let's give this project a try!

It is a really great idea!

[rommelnc]

do I need to say YES? or just the THUMB UP is ok?

Anyway, here I go:

YES + THUMB UP

Games published by our own members! Check 'em out!

[rommelnc]

This project's goal is to implement Artificial Inteligence algorithms,
the API, and make them available for any platform but especially for
mobile devices.
 
The AI techniques that will be implemented are:
 -Finite State Machine
 -Fuzzy State Machine
 -Scripts
 -Agents
 -Neural Network
 -Decision Tree
 -Bayesian Network
 -Etc.
 
The main objective is to superpass known problems with the mobile
devices limitations and allow the community to develop better and more
inteligent mobile games.
--------
The project members can be contacted by the community on the following
e-mail: estudosemia@gmail.com
--------
As the implementation part of the project haven't started yet, we are
not requiring any extra participation. But, at the time of the first
release, we will be very pleased to have members of the community
testing our stuff.
-------
To the testers, is good not only to have J2SE understanding, but a
little experience on the AI techniques that we will be implementing in
order to understand our approaches and suggest other approaches too.
In the future, a nice level of knowledgment in J2ME will be required
also.

[rommelnc]

I can't see the image in the first message! What's up?

[tom]

+1

[Tiagobc]

I think that this project is really good!!!!!!!!!     

[carlosbpf]

Really cool J2ME gamming programming artificial inteligence API =)!!

[andersonfer]

it seems to be a nice project!  =)

[Riven]

What we got here is account abuse. Kick kick kick

Games published by our own members! Check 'em out!

[blahblahblahh]

vote no, because:

1. the original proposal seems complete pie-in-the-sky, no-idea-what-we-re-doing. c.f. gamedev.net for examples of people who think AI is "really cool" and that they can just start hacking on it and expect to get somewhere if they try hard enough (no, AI is not that well understood; you could hack on it for 10 years and still have nothing to show for it)

2. the account abuse. The first one was suspicious. The others are just taking the piss

EDIT: PS if I were a mod on this, I'd be tgaking careful look at those IP's, and number 2 above I wouldn't be having to cast aspersions that might be totally unfair. I cant do that on this cat, so have to just guess

[rommelnc]

Hi everyone!

Please have patience and read this message.

I would like to answer suspicious about account abuse.

First, I would really like that you check the IP. It is really a good ideia I have no problem with that. Second, everybody has to start posting some day. I am the "owner" of the project, and I am not afraid to say so (you can see that the nickname is the same). If I can not vote please take just disconsider it. I am not familiar with the rules, so please forgive me.

I know that AI is a hard subject. That is why I have been studying it for so long (it has been more than 4 years now). To be difficult is not a reason to disapprove an aproach trying to take out the best of it. I suggest you check the project UnBBayes is SF.net (it works with Bayes Network, Influence Diagram and Multiple Seccioned Bayesian Network and learning techiniches for it - all implemented, testing and working). You will see that it is a relativelly old project and it is still active. There is one game, futsim (soccer simulation), that uses our UnBBayes API. We have also the UnBMiner that works with Neural Network, Decision Trees and some other things. Our APIs were also used in a program for recognizing digital prints for our Federal Police. Our teachers (Phd.s) oriented a group in the same technology we developed and this group developed their own for a Bank (Brazil Bank - www.bb.com.br) for detecting credit card fraud, and they had a really good result with it (millions saved a month!). We have several articles published and so on...

Finally, I ask you to please check the information I am passing to you guys. Please, send our teacher in University of Brasilia (Brazil's capital) an email, mladeira@cic.unb.br - Department of Computer Science, his name is Marcelo Ladeira - homepage: http://www.cic.unb.br/docentes/ladeira/ladeira.html.

Hope you guys ask more questions so I can answer and clear doubts out. 

I am trying to use the java.net site, as I always worked with java, but if I do not have the support of this community I will stay with SF.net as we did with our other projects.

Thanks,
Rommel Novaes Carvalho
rommelnc@centrodeinteligencia.com.br

[rommelnc]

Here is the link for the soccer game: www.futsim.com.br!

[mj2005]

the  project´s goal seems to be really intersting and i would like to know more about it

1 - what  is the name of the project?
2 - the source code will be disponible under what license?
3 - the group has already implemented (or is implementing) sometinhg?
4 -  the material that  the group is using for study AI  is disponible on the Web?

Thanks and good luck with the project

Marcelo Bezerra
Empresa Jr. de Computação - www.cjr.cic.unb.br

[mj2005]

by the way .... + 1   

[rommelnc]

Hi, thanks for the interest.

The name is UnBGames (name of our University + Games).

As sugested by the java.net gaming community we are going to put it under BSD license.

We do not have code yet, but we will soon have. We already have UML diagrams for Mealy and Moore FSM, Fuzzy Logic and Neural Network (we are finishing GA soon).

Last semester we wrote 100 pages related to FSM, NN, GA, Fuzzy Logic, history of games and games categories with AI use. It is only in Portuguese, but we plan on writing an english version.

We are going to put in our page all links we used to study and write our stuff.

Thanks,
Rommel Novaes Carvalho
55-61-8111-7048
rommelnc@centrodeinteligencia.com.br

[Jeff]

Okay, lets clarify the issues here...

I assume what BBB is referring to is the large number of "noobs" voting on this, and that two of the "noobs" appear to have been created withn 15 minutes of each other.

While that is *slightly* suspicious I can see a perfectly reasonable explaination, myself.  The poster has always maintained that this was a team project and I don't know of anything in the rules that says he cant encourage his team members to join and vote.  In fact, as that could enlarge the community, I think there is some potential value in such things.

Im not sure IP addresses would mean much either since, if they are a team/group of friends, it is perfectly reasonable that two or more could have registered from the same system.

I also don't remember a rule that says the poster can't vote for his own project, but I'll double check and if there is I will discount that one vote.

HOWEVER since protocol issues have been raised I would like to hold this over one more week.  In that time I would like the JGO board to look at this and make some decisions on (a) how to move forward on this one and (b) what if any changes in the rules they would like to make for the future.

JGO Temporary Administrator
Jeff Kesselman

[rommelnc]

While that is *slightly* suspicious I can see a perfectly reasonable explaination, myself.  The poster has always maintained that this was a team project and I don't know of anything in the rules that says he cant encourage his team members to join and vote.  In fact, as that could enlarge the community, I think there is some potential value in such things.

That is exactally what happened. One vote is mine and there are 3 more from members of the project (there is still one member that did not vote, yet! ). They are different people and I assumed they could vote here with no problems. I understand why some people are suspicious, but I garantee that they are not the same person. That is why I have been posting here trying to clarify things.

I also sent e-mails to students of Computer Science in UnB (University of Brasilia), so maybe there is going to have more votes from n00b. I hope there is no problem with that, after all they are computer and game lovers!

Thanks again,
Rommel Novaes Carvalho
55-61-8111-7048
rommelnc@centrodeinteligencia.com.br

[Jeff]

It is my understanding that the JGO board has decided that this project will not be approved.

If I am wrong, someone from the board should contact me directly.

[rommelnc]

Is there any way you can explain me why? Anyway, thanks for the chance. I will look for someone who can support us.

Thanks,
Rommel Novaes Carvalho
rommelnc@centrodeinteligencia.com.br

[blahblahblahh]

Is there any way you can explain me why? Anyway, thanks for the chance. I will look for someone who can support us.

Thanks,
Rommel Novaes Carvalho
rommelnc@centrodeinteligencia.com.br

Try http://sf.net ?