Propose your ideas

[cezar_1]

Well, it's always good to know the priorities. Taking them into consideration I think I will take an interest in "Dynamic attributes and statistics" idea. I consider this as a very interesting and necessary feature and I am sure I can deal with the related tasks. I will discuss the idea in a proper topic. But firstly I will need to ponder upon it of course

By the way, sorry the lack of my presentation. I will introduce myself soon.

[lcjunior]

Hello,

My name is Luiz Carlos Ferreira Junior and I'm from Brazil and I very
interested by the project Data Laboratory.

I'm currently the fourth year of Computer Engineering and also do
training in the area of software development using various programming
languages, as well as design standards and their proper documentation.

The project caught my attention because its goal is also something
that I believe and nothing better than to develop something that you
really believe and believe.

Another thing is that the knowledge required for their implementation
is knowledge that I have been searching for throughout my life in the
area of software and I want to learn more and more.

Well, I am very excited and really want to participate in GSoC is my
first participation and work on something that improving people's
lives as well as the free software community.

Then there was the possibility that I have a proposal for this
project. If you can have a model or template to follow.

Thank you.

[Dhanushka]

hi

I'm Dhanushka from University of Moratuwa Sri Lanka. I'm really interested in the "Web-based network visualization engine with WebGL" idea.
I have worked with javascript for a long time. I also have a big interest in 3d visualization. I want to get involve in that part. I'm currently studding OpenGl with c++. I think this is a big opportunity for me to get involved in a realistic project and get experience.

thank you.

[admin]

Hi, this is not the good thread for presenting yourself. Please use the proposer discussion and let this one for real idea proposal.

[cklee]

I dug up this page while I was googling Gephi + API design.

I think this is an archived wikispace from last year's GSoC, but I thought it'd be useful here.
It's just a quick and dirty intro to Netbeans, Gephi framework + API, etc.

http://gephigsoc.wikispaces.com/

[admin]

Hehe, but take care that APIs have evolved

[mbastian]

Hehe, but take care that APIs have evolved

Quite a lot indeed, everyhting should have been moved to the Gephi Wiki.

[luizribeiro]

I would like to propose a new idea for GSoC 2011 and hopefully get some feedback from the community. I've already talked with some people on Gephi's IRC channel about it, but I also would like to discuss it with possible mentors and other students.

I don't know if you're familiar with Tinkerpop's Blueprints, but basically it's a property graph model interface (think JDBC, but for property graph databases). There are some useful graph tools that work with Blueprints and, by having a Gephi Blueprints implementation, we could use all these tools with Gephi.

By having a Blueprints implementation, users could use the Gremlin graph traversal language to query the graphs created within Gephi, enabling a powerful tool to work with the structures and test new algorithms.

Also, we could have a MuTanT console inside Gephi. A MuTanT console would allow access to the graphs created with Gephi in any scripting language that has a ScriptEngine compatible with JSR 223 (Ruby, Gremlin and Groovy, to name a few).

One of the main advantages of a MuTanT console is that you can change the current ScriptEngine without losing the variable pool. Or, in other words, you could import the graph into a variable, run some source code in Groovy, change the language to Gremlin and do a graph traversal, without losing the global variables on the variable pool, for instance. You can see a MuTanT example session like this here.

Since the MuTanT console would be a powerful tool to experiment with algorithms on the graph structures created within Gephi, this feature would certainly bring a new kind of users to Gephi: people interested not only in graph visualization and manipulation, but also graph algorithms. For example, one could experiment a new flow algorithm to solve some optimization problem with a graph created in Gephi in just a few minutes. I'm sure that many of the current Gephi's users also would like to have a tool to experiment with algorithms and querying graphs under Gephi.

What do you think about this idea?

[mbastian]

Hi Luiz,

Thank you very much for digging into this and sharing your thoughts.

You are making a very good point.

We are indeed familiar with Blueprints, great project. Implementing Blueprints in Gephi is possible, we already looked at possible blockers and didn't find any major. Several things remain unclear though. If I understand well, Gephi could be both an implementation and an ouplementation, right?