wouwlabs

jeraman

2010-03-11T21:15:51+00:00

Marvim Gainsbug é músico, cantor e compositor, criado em 2009. Suas principais influências são Serge Gainsbourg, Bob Dylan, o Repentista Nordestino, Leonard Cohen, Joni Mitchell, Alan Turing, Deep Blue, HAL, Wintermute e Marvin – o Andróide Paranóico.

Marvim Gainsbug é um software que atua baseado no Twitter, programado com o objetivo de compor e executar canções, com letra e música, em tempo-real.

o tweets se transformam em versos que serão interpretados pela marcante voz de Marvim. a melodia, a harmonia e o ritmo são diretamente influenciados pelas palavras que compoem os “versos”. para tanto, basta que o visitante diga em um microfone qual a temática da sua canção.

interface do Marvim Gainsbug

a interação proposta para esta versão do Marvim Gainsbug pode ser descrita como uma experiência audiovisual que evoca a impressão de interagir com um repentista, que através de seus sentidos (é capaz de “ver”, “falar” e “escutar”) e sensibilidade (de criar música), é capaz de criar uma canção, letra e música, de maneira “improvisada”, em tempo real.

criado em 2009, em parceria com Filipe Calegario, o projeto foi construído em Processing, utilizando o Twitter4j para lidar com a pesquisa no Twitter, a Sphinx4 para o reconhecimento de discurso (Speech Recognition) e a FreeTTS para síntese de voz. a interação foi viabilizada por um microfone posicionado em frente a projeção, por sua vez, utilizada como saída de vídeo para o projeto, e caixas de som, responsáveis pelo retorno sonoro do trabalho.

Marvim Gainsbug em ação

escutar músicas, ver fotos e maiores informações, acesse o site do projeto:

myspace.com/marvimgainsbug

maiores informações

[1] sobre o Marvim Gainsbug
[2] sobre o processing
[3] sobre o Twitter4j
[4] sobre o Sphinx4
[5] sobre o FreeTTS

Tagged: marvim gainsbug, processing, projects

jeraman

2010-03-07T18:20:24+00:00

os últimos dias andam bem corridos, mas produtivos… eis algumas novidades:

LaboCA

vai rolar entre os dias 22 a 26 de Maio, no Memorial Chico Science, como parte integrante da Semana Chico Science, a primeira edição do LaboCA, Laboratório de Computação e Artes, criado recentemente por mim, Jarbas Jácome e Ricardo Brazileiro.

o LaboCA é um projeto de laboratórios experimentais nômades com o objetivo de ensinar e pesquisar sobre o uso da computação para fins artísticos. ao final de cada laboratório esperamos ter como resultado objetos de arte de natureza computacional (softwares e/ou hardwares) produzidos coletivamente pelos participantes.

maiores informações estarão disponíveis em breve do site do projeto, laboca.org.br, ou no site da Semana Chico Science.

II Encontro Livre

um ano se passou desde da excelente experiência que tive em participar do I Encontro Livre pois bem… desde então, o evento amadureceu, cresceu e vai ocorrer novamente esse ano, entre os dias 10 e 12 de Março, na Livraria Cultura. o evento, organizado pelos brothers gatonildes seucolega e célia, visa promover o diálogo entre a cultura livre e assuntos que permeiam sua construção (tecnologia, meio ambiente, propriedade intelectual, comunicação livre, …).

maiores informações no site do evento, encontrolivre.org.

FILE

essa é a última semana pra quem quiser submeter trabalhos para o FILE, o Festival Internacional de Linguagem Eletrônica. este ano, uma novidade: ao inscreverem projetos para participar do evento, os artistas estarão também concorrendo ao prêmio FILE PRIX LUX… maiores sinformações no site!

o FILE é também a principal razão da correria desses dias. espero nos próximos posts falar um pouco dos projetos que vamos submeter por aqui…

até a próxima!

Tagged: others

Imagem10

2010-02-08T20:39:27+00:00

passado o teste inicial, decidi pesquisar sobre o ponto que realmente me interessava: usar o openframeworks no desenvolvimento de aplicativos para o iPhone OS! essa possibilidade é fuderosa e vem sido utilizada com sucesso por várias pessoas por aí, incluindo cão-chupador-de-mangas Memo, autor do Zoetrope, MSA Remote, Gold Dust, entre outros…

o resultado desse experimento foi o seurat: um protótipo de aplicativo para a plataforma iPhone OS, inspirado no Seurat vai à Rússia.

nele, o programa baseia-se numa foto escolhida previamente, com o objetivo de desenhar aleatoriamente uma série de pontos visando reconstituir a foto original sob uma perspectiva impressionista. o resultado é construído pelo usuário, que interfere no processo tocando na tela, direcionando assim o local onde os pontos surgirão.

o “pulo do gato” que permitiu usar o openframeworks dentro do iPhone OS é o ofxiPhone, criado por, dentre outros, o próprio Memo…

openframeworks no iPhone!

o ofxiPhone é um addon para versão 006 do openframeworks, que lhe permite desenvolver aplicativos para iPhone OS da mesma forma como se programaria para um desktop normal, em openframeworks C/C++! isso implica que, todas as facilidades trazidas pelo oF passam a ficar disponíveis também para o iPhone; a lógica de programação, a biblioteca, os addons… obviamente, essa semelhança esbarra na limitação de hardware do dispositivo…

todas as informações relativas a como realizar essa integração (downloads, instalações, configurações), que por sinal é extremamente fácil, podem ser encontradas no site do openframeworks… existem exemplos mostrando como usar o acelerômetro e o multitoque, o que será bem fácil para quem já trabalha com oF!

além disso, recomendo bastante a leitura do artigo escrito por Memo, sobre o addon, antes de meter a mão na massa. para os usuários mais avançados, vale a pena também dar uma sacada no guia escrito por Zach Cage…

maiores informações

[1] sobre o openframeworks
[2] sobre o iphone sdk
[3] sobre o ofxiphone

Tagged: iphone, of, projects

Imagem2

2010-02-08T19:10:27+00:00

motivado pelos diversos aplicativos interessantes que andam saindo pro iPhone OS por aí, baixei neste fim de semana o kit de desenvolvedor (iPhone SDK versão 3.0), pra brincar um pouco e ver como é que seria desenvolver para essa plataforma.

antes de tudo; pra desenvolver oficialmente, eles lhe obrigam a usar o Mac OS e a IDE oficial da Apple; o xCode. parece que existem formas de burlar isso daí, mas não foi o o foco dessa pesquisa… pra os que seguem pela primeira opção, existe uma documentação farta no próprio site da apple, auxiliando em diversas etapas do processo de desenvolvimento.

tendo isso em mente, o primeiro passo é baixar a SDK. além de contar com todas as bibliotecas necessárias pro desenvolvimento, ela conta também com um simulador do hardware, pra testar as coisas antes de ir pro dispositivo físico.

simulador rodando minha primeira aplicação!

instalar e configurar é um processo simples (igualmente bem documentado no site da apple). a dor de cabeça maior ficou por aprender a sintaxe básica de Objective-C, a linguagem – baseada em C – usada para programar na plataforma, e relembrar um pouco de Design Patterns (que por sua vez me relembrou os não-tão-saudosos tempos do Centro de Informática…) amplamente utilizado na arquitetura das aplicações dessa plataforma.

pra dar meus primeiros passos, segui um tutorial bem legal: “Your First iPhone Application”. além de surpeficial, achei em alguns momentos a coisa meio difícil de ententer, mas deu pra sair sem maiores problemas.

o resultado é uma aplicação que contém um label cujo conteúdo é definido por um text area, no momento que o usuário aperta o botão “Hello iPhone SDK”… bem útil, não? a interface pode ser conferida na imagem acima! se quiseres algo mais aprofundado que isso, podes conferir o iPhone Development Guide.

como saldo final, fica um ponto negativo: pagar por uma licença pra se tornar um desenvolvedor oficial… isso mesmo! pra se tornar um desenvolverdor-oficial-apple-foda-come-cuzes e assim passar seus aplicativos do simulador pra o aparelho, você terá que pagar a Apple e, além disso, o valor é uma bagatela simbólica no valor de 99$!!!

portanto, se sonhas em ficar um dia milionário vendendo aplicativos na Apple Store, saibas que primeiro terás que ter 99$ pra comprar uma licença de desenvolvedor oficial… ¬¬

o próximo passo, testar o SDK junto com o Openframeworks, fica pra o próximo post!

maiores informações

[1] sobre o iphone os
[2] tutorial: construindo minha primeira aplicação para iPhone
[3] iPhone Development Guide
[4] overview da linguagem Objective-C

Tagged: estudos, iphone

filipecalegario

2010-01-31T20:42:55+00:00

Olympiapark, München

2010-01-29T20:05:58+00:00

filipecalegario

2010-01-26T20:08:35+00:00

filipecalegario

2010-01-24T20:05:48+00:00

filipecalegario

2010-01-19T19:56:03+00:00

Since the last post a lot of work has been done.

On the 4th of February all the teams will be together in London to present the results and party! =D

If you want to have more details, you can access our site on Nokia PUSH official blog:

http://blogs.nokia.com/pushn900/category/sketch-your-world/

Below you can see some videos about the project:

Imagem2

2009-12-09T16:37:08+00:00

apesar de não ser nenhum especialista no tema, mesmo gostando muito de criatividade/métodos/técnicas criativas, fui convidado recentemente pela professora Carina Alves para fazermos uma dinâmica sobre Criatividade na Concepção de Softwares, às 8:15 do dia 10 de Dezembro, lá no Centro de Informática, durante o WIn – o I Workshop de Inovação.

achei a oportunidade uma experiência interessante: serviu pra tentar sintetizar e organizar todas as informações que eu tinha na cabeça sobre o tema. assim, além da dinâmica propriamente dita, optei por criar uma pequena introdução, que fizesse as pessoas pensarem sobre criatividade de uma nova forma: não mais como algo místico e inalcançável, mas como uma competência inata ao ser humano, a ser trabalhada e desenvolvida, além de imprescindível nos dias de hoje.

os slides podem ser conferidos logo abaixo. quem sabe eles não acabam sendo úteis pra alguma outra pessoa?

ps: pra’queles que se interessam pelo tema, deixo uma dica: o zé maria precisa de idéias. apesar de andar meio parado, o grupo tem como objetivo ser um espaço aberto para discutir criatividade… caso deseje participar, o endereço do grupo é o:

http://groups.google.com.br/group/ze-maria-precisa-de-ideias/

nos encontramos por lá!

Tagged: criatividade, lectures

mosaico

2009-12-07T00:41:16+00:00

‘‘onde está você na República?’ – projeto efígie – é uma instalação interativa na qual a efígie da república, a “personificação” do regime, é reconstruída a partir de nova ótica: formada a partir de fotos de cidadões comuns, sendo tais fotos retiradas no próprio no local da instalação.

efigie project

o projeto efígie foi desenvolvido em dois módulos: a máquina e o mosaico. o módulo máquina, desenvolvido em C++ com openframeworks, é responsável por detectar a presença de pessoas em frente a câmera e retirar as fotografias. dessa forma, as fotos retiradas são enviadas através de uma rede wireless para uma segundo computador, que executa o módulo de mosaico.

esse segundo módulo, desenvolvido em processing, é então responsável por fazer a montagem da releitura da efígie, exibindo o resultado final em dois projetores. esse processo é feito seguindo uma linha de animação que pode ser conferida abaixo:

módulo mosaico

com conceito criado pelo artista plástico Itamar Morgado e eu, o trabalho está atualmente exposto no Espaço Cícero Dias do Museu do Estado de Pernambuco, em Recife, como parte integrante de uma exposição comemorativa dos 120 anos da República no Brasil, organizada pelo Governo do Estado, que ficará em cartaz até o dia 7 de Fevereiro.

também está rolando no local um trabalho bem legal de Gabriel Furtado, do Media Sana…. coisa finíssima! quem quiser dar uma conferida, é só dar uma passada por lá! pra quem não pode e ficou curioso, aí vão algumas fotos:

maiores informações

[1] sobre a exposição comemorativa dos 120 anos da República
[2] sobre a Efígie da República
[3] sobre Processing
[4] sobre Openframeworks
[5] sobre o Governo do estado de Pernambuco

Tagged: efigie, of, processing, projects

post-monera

2009-11-29T00:45:50+00:00

M.O.N.E.R.A. is a new way to visualize Twitter information using an organic metaphor: each tweet is represented by a bacteria inside an ecosystem and each species represents a topic. New information is acquired dynamically by Twitter API which makes the environment live and unpredictable.

The project was developed as part of the Information Visualization class lectured by Prof. Gudrun Klinker, TU München [1]. It was built upon the ideas of Carolin Horn and Florian Jenett in the project Anymails [2] (www.carohorn.de/anymails) and Ben Fry in the work Organic Information Design [3]. The illustrations of the creatures come from the book Kunstformen der Natur by Ernst Haeckel, about which I posted a month ago. The project was programmed using Processing [4] and it is using twitter4j to access Twitter API [5]. The engine is based on Nature of Code by Daniel Shiffman [6].

You can see the version it will be presented on the show room “ITüpferl” from Fakultät für Mathematik und Informatik at TU München here:

http://www.filipecalegario.com/monera_itupferl

Or you can see the version on OpenProcessing site:

http://www.openprocessing.org/visuals/?visualID=6261

References:

[1] Information Visualization SS2009 - campar.in.tum.de/Chair/TeachingSs09InfoVis

[2] Anymails - www.carohorn.de/anymails

[3] Organic Information Design - benfry.com/organic

[4] Processing - processing.org

[5] Twitter4J - yusuke.homeip.net/twitter4j

[6] Nature of Code - shiffman.net/teaching/nature

Imagem1

2009-11-27T17:03:29+00:00

a realidade diminuída

se você também já encheu o saco desses projetos de “Realidade Aumentada” com marcadores, esse trabalho é pra você: o “Realidade Diminuída”, por Mozart Santos, Jarbas Jacome e eu, mescla a tecnologia com um outro tipo de “Realidade Virtual” bem mais antiga e clássica, os espelhos, pra tratar justamente sobre… Realidade Aumentada! o resultado pode ser experimentado no site abaixo:

http://jeraman.info/projects/ra-diminuida/

desenvolvido em flash e html, o trabalho foi possível graças a duas tecnologias para Flash bem interessantes; o PaperVision3D e o Flartoolkit. o primeiro é uma biblioteca para lidar com 3D (cuidando da parte de exibição), cumprindo esse propósito de maneira formidável. o segundo é uma extensão da biblioteca C++ Artoolkit, responsável por toda a parte de Realidade Aumentada, que, pra minha surpresa, também funciona muito bem – apesar de não ser tão bem documentada.

o código fonte está disponível aqui e é livre para ser estudado/modificado/pregado-na-parede/que-quer-que-seja.

a idéia surgiu num encontro lá no Memorial Chico Science com Mozart Santos, Jarbas Jacome, eu e o Fábio Rafael… por sinal, um encontro bastante produtivo!

maiores informações
[1] sobre o Flash
[2] sobre PaperVision3D
[3] sobre o Flartoolkit
[4] sobre o Artoolkit

Tagged: flartoolkit, flash, papervision3d, projects, realidade diminuida

jeraman

2009-11-25T13:53:07+00:00

“seja qual lado se corra,
sempre se é
um pedaço pequeno
da hora que já foi.”

“espelho recursivo” é um trabalho no qual o vídeo captado pela webcam sofre uma releitura em tempo-real, passando a ser composto por frames antigos, que passam a exercer o papel de “pixels” do vídeo original. dessa forma, tudo o que é filmado pela webcam acaba sendo mostrado tal como um espelho, mas sob uma nova ótica: o do presente como a composição de fragmentos do passado.

inspirado no Estereoscopia, de André Parente, o trabalho foi desenvolvido em C++, usando openframeworks, sendo baseado na máquina de um outro trabalho sobre o qual espero falar muito em breve: matéria pra um outro post.

abaixo segue alguns screenshots com imagem em boa resolução. espero que gostem!

maiores informações
[1] sobre openframeworks
[2] sobre André Parente

Tagged: espelho recursivo, of, projects

jeraman

2009-11-21T14:55:43+00:00

depois de muitos testes e desenvolvimento, chegamos ao lançamento do novo release do projeto VI (Vitrine Interativa), já comentado aqui mesmo no blogue, que estou desenvolvendo em parceria com o pessoal da Code Media.

inspirado em cenas do filme “Minority Report“, de Steven Spielberg e desenvolvido em C++, utilizando openframeworks, a tecnologia permite que uma pessoa por vez interaja com a vitrine a distância, usando uma abordagem touchless (sem toques). desta forma, basta que o usuário aponte para qual dos quadrantes ele deseja selecionar e a tecnologia faz toda a mágica…

um ponto bastante interessante do projeto VI (tanto do release “minority report” quando o do “menino”) é que, devido a sua arquitetura, ele consegue abstrair em qual linguagem será desenvolvida a aplicação final! dessa forma, a customização do software a ser exibido fica extremamente fácil de ser realizada. no demo do vídeo acima, por exemplo, a aplicação demo (quadrantes coloridos) foi escrita em ActionScript3 / Flash…

bem, é isso… final de ano tá chegando, mas a produção continua!

maiores informações
[1] sobre o projeto VI
[2] sobre a Code Media
[3] sobre openframeworks

Tagged: flash, of, projects, vitrine interativa

Sumot - Prezi Presentation

2009-11-15T20:36:18+00:00

A few weeks ago, the Winter Semester 2009/2010 at Technische Universität München started and the hard work began again! I’m attending the Practical Course of Augmented Reality Applications and the challenge is to develop a game using Lego Mindstorm NXT and Augmented Reality.

Tomorrow will be the first presentation where we will present the inital idea of the game. There is no many details yet, but you can have a look and get some draft ideas. You can take a look…

The presentation was made using Prezi and you can see it here:

http://prezi.com/z2tsu4rc4ojh/

That’s it! =D

filipecalegario

2009-11-10T16:47:16+00:00

Yesterday we had great news: our Nokia PUSH project ‘Sketch Your World’ was selected to be implemented!

Here is the project video:

And here is the PDF.

Now we have to work hard to give life to this beautiful idea! (OH MY GOD!) =D

More information: Nokia PUSH Official blog, Jeez post.

“You are a PUSH N900 winner!”

2009-11-09T20:56:18+00:00

… as the subject of the email I received today said. I was so excited when Mat Dobson called me earlier today that I’m still putting my ideas back in order! :)

It’s official! Sketch your World is now one the Nokia’s PUSH N900 winner.

Don’t remember the project?! Check the previous post and/or the PDF and video submitted:

You can follow up the project development on the Push blog.

And now we must run for finishing it up!

jeez

IMG_3005

2009-11-04T12:51:47+00:00

engrenagem em ação no Continuum

recentemente, em parceria com Ricardo Brazileiro e Seu Fábio, participei de uma experiência bem instigante, tensa e divertida: a construção da Engrenagem.

a Engrenagem é uma máquina criada por Maíra Sala e Bruno Vianna que traz a tecnologia multitoque para um domínio bem particular: o Live Cinema (modalidade do cinema que acontece “ao vivo”, sendo todas as cenas editadas em tempo real, na hora da exibição, pelo diretor para o público). o projeto já esteve outra vez no Recife, com a exibição do filme Ressaca pelo próprio Bruno Vianna, no início desse ano lá na Fundação Joaquim Nambuco. a aplicação foi desenvolvida com o processing e o pd.

aprendemos bastante com a essa construção, que ficou – palavras de Seu Fábio – profissional demais. o trabalho ficou exposto no I Festival de Arte e Tecnologia do Recife, o Continuum, já comentado aqui mesmo no blogue. pra quem não pôde dar uma conferida, segue algumas fotos!

======

mudando um pouco de assunto, mas permanecendo no tópico multitoque, mais uma novidade: com a finalidade de juntar o pessoal que trabalha/tem interesse em tecnologias multitoques no Brasil, criei na última semana uma lista de discussão para que possamos trocar idéias, tirar dúvidas, compartilhar experiências, e por aí vai… o endereço da lista é:

http://groups.google.com.br/group/mt-br

para participar, basta se inscrever… espero que a lista seja proveitosa para todos!
nos encontramos por lá!

maiores informações

[1] sobre a comunidade Multitoque Brasil
[2] sobre o Continuum
[3] sobre o Ressaca e a Engrenagem

Tagged: brasil, engrenagem, multitouch, pd, projects

Qt Mobile Demos

2009-11-03T19:47:54+00:00

openBossa has released Qt Mobile Demos, a pack of fancy demos for showing off what Qt 4.6 can do!

Until now, there are four of them: MyBudget, ShoppingList, Weather and HyperUI.
You can check out their source code (master branch is for S60, maemo branch is for maemo), or you can just download S60 and/or Maemo5 Packages and try them out. There still some work needed in a place or another, but the overall status is very impressive: they run really smooth on both devices. MyBudget stills using QTimeLine instead of the Qt Animation API, but this “porting” is already in progress. The maemo packages have been compiled using Qt-X11-Maemo repository, but we also uploaded these packages for you. :)

For the lazy ones, a few screenshots:

HyperUI

Weather

ShoppingList

MyBudget

As you can see, each demo has its own conceptual idea. Our designers managed to do a great work! Patricia Montenegro, Ian Moreira and Tamara Baia made a 45″ video presenting the whole design process of the Weather demo. Patricia (patifa) also wrote a very complete post about this particular demo.

Changing the weather from Ian Moreira on Vimeo.

Let’s hope that they prepare some more nice videos about the other demos!

It was really fun to work on this. Congrats to all teams involved in the project!

so Qtie! :)

jeez

filipecalegario

2009-11-02T05:04:22+00:00

We didn’t win, WE WON (http://blogs.nokia.com/pushn900/index.php/2009/11/09/if-we-could-draw-your-attention-for-a-moment/)! but we tried hard. That’s the point with everything. =D

More details: jeez’s post and Nokia PUSH official blog.

máquina soma 2

2009-10-29T18:32:43+00:00

trecho do processo3_soma

recentemente, conheci através do mestre das gambiarras low tech Ricardo Brazileiro, o dançarino Marco Bonachela. através desse contato, surgiu o convite para integrar o projeto soma, desenvolvendo interfaces interativas para a próximo espetáculo do grupo; o processo3_soma, para o 14º Festival Internacional de Dança do Recife.

surgido em 2007, soma é um projeto muito interessante e ambicioso. propõe um diálogo entre áreas completamente distintas: a dança a música e a tecnologia. para isso, conta com dois músicos (Hugo Carneiro e Henrique Lima), um dançarino (Marco Bonachela) e um programador (eu), propondo processos contínuos de construção, numa perspectiva ligada ao work in progress.

Marco Bonachela, Henrique Lima e Hugo Carneiro

a estrutura básica das produções do projeto funciona da seguinte forma: os sons criados pelos músicos e sapateador são capturados para um computador, onde se tornam parâmetros para a construção em tempo real de vídeos. a interface se torna possível através da criação de efeitos no Pure Data (vulgo Pd, já muito comentado aqui mesmo, neste blogue), manipulado também ao vivo durante a apresentação.

para tanto, desenvolvi um patch para o pd com alguns efeitos audiovisuais interativos (loopers, trackers…) para a performance, a máquina soma, cujo resultado final achei bastante interessante. sua construção foi baseada nos 2 quadros que compõem o evento, cada um composto dois 4 temas diferentes. dessa forma, cada tema possui um efeito específico. na performance, também usei um pequeno módulo de tracking (pra saber onde o dançarino estava no palco), desenvolvido em openframeworks (tal módulo por sinal, é a base de um projeto ainda em desenvolvimento… em breve mais detalhes!).

a máquina soma no Teatro Hermilo Borba

pra aqueles que desejam experimentar um pouco dela, a máquina soma é livre e está disponível para download aqui (por conta do tamanho sem nenhum video ou som. para mudar isso é só adicionar as novas mídias dentro dos objetos “loaders” localizados dentro de cada tema). espero realmente que ela seja útil de alguma forma para outras pessoas que desejem brincar, estudar, realizar outras performances, não fazer nada e por aí vai… =P

neste dia 28 de outubro, às 20h, rolou uma apresentação do processo3_soma lá no Teatro Hermilo Borba. apesar dos imprevistos e dos problemas, achei pessoalmente a apresentação final bem legal e estou muito feliz com a experiência: parabéns a todos que participaram do processo! as fotos podem ser conferidas aqui, no flickr do Festival. que venham as próximas!

maiores informações

[1] 14º Festival Internacional de Dança do Recife
[2] sobre o Pure Data
[3] sobre o Openframworks

Tagged: dance, music, pd, projects

Kunstformen der Natur

2009-10-21T19:36:11+00:00

I while ago I read in a

what was defined by Manuel Lima [http://www.visualcomplexity.com/vc/blog/?author=1] as “outstanding work of nineteen century German biologist Ernst Haeckel”. I got quite interested, primo because I find it very attracting the art that exits behind the nature forms, secondo it is quite nice to see works from the past.

Seurat vai à Rússia… [http://jeraman.wordpress.com/2009/09/05/seurat-vai-a-russia/]

where jeraman got inspired by a color photography of the beginning of 20th century by Prokudin-Gorskii

Kunstformen der Natur

I while ago I read two posts [subblue's post] [visualcomplexity's post] discussing an illustrated book of 1904 which was defined by Manuel Lima as “an outstanding work of nineteen century German biologist Ernst Haeckel”. The name of the book is …

… Kunstformen der Natur (Artforms of Nature), its copyright has just expired and you can admire it on Wikimedia Commons [Kunstformen der Natur] or download it here. The pdf file has more than 260 megas with high-res amazing pictures.

I got quite interested, primo because I find very attracting the art that exists behind the nature forms, secondo it is quite nice to see works from the past, as Visions de l’an 2000 (Visions of the year 2000) or Seurat vai à Rússia…, where jeraman got inspired by the colored photography by Prokudin-Gorskii of the beginning of 20th century.

After hours and hours admiring the wonderful collection of illustrations, I had the idea to select some pictures and transform them in sprites to be used in artificial life simulation softwares or games. I think it would give an impressive and beautiful result for those systems.

So, I list below the steps I followed to transform the images:

Copied the image from the book
Pasted in Adobe Flash
Modify -> Bitmap -> Trace bitmap
Most used threshold 100, some cases 50 and Color limit 25
Removed the background and unwanted stuff
Exported as PNG

As a result, I present the pictures in the gallery:

Qt Labs Americas no Latinoware

2009-10-20T14:25:48+00:00

Entre 22 e 24 de Outubro estará acontecendo o Latinoware 2009, no Parque Tecnológico de Itaipu – Foz do Iguaçu.

Nós estaremos por lá! Eu e Anselmo Melo vamos ministrar um mini-curso de Qt 4.6 no sábado, 24/10. Teremos também palestras de Artur Souza, sobre o Plasma Netbook, Caio Marcelo e Eduardo Fleury falarão sobre Qt Layouts, Ana Cecília (aninha) vai falar sobre Usabilidade em OSS usando seu GSoC como estudo de caso e Kenneth vai falar sobre o Webkit.

Caso queira entrar em contato, basta nos procurar nas referidas palestras ou no stand do INdT que estará em algum lugar do evento. :)

Nos encontramos por lá!

jeez

Nokia Push N900

2009-10-19T21:19:11+00:00

Nokia has launched a hacking/modding contest for N900 called Push.
The main idea is based on “how would you connect something that
you love to a Nokia N900 ?”. There will be 5 winners, who will
receive funding and a N900 to make their projects become real,
and after that all projects will be displayed on several events
around Europe, on Nokia Flagship Stores. Check their FAQ for more info!

At first, the whole wouwlabs group decided to participate as a team
(a team could have up to 5 members). We made some meetings and
brainstorms and came up with several ideas. And good ideas!
But in the end, just me, annieC and calega managed to have enough
“spare” time to finish everything. On the last week we invited
Patricia Montenegro (patifa), a brazillian designer, to
enter the team and help us out!

Well, I’m very happy with the result of all submitted projects.
Below you can checkout the final documents and video for each of them:

- Pignocchio Bank: A re-design of the classical Piggy banks!
Document: Pignocchio Bank (pdf)
Video: http://www.youtube.com/watch?v=A0Zwhtf0ThA

- Sketch your World!: Automatic draws on a Etch-a-Sketch using a N900!
Document: Sketch your World! (pdf)
Video: http://www.youtube.com/watch?v=EYDyNwXqjjg

- N9P0V: A Persistence of Vision gadget using a N900 as drawing, animating and message writing tool!
Document: N9P0V (pdf)
Video: we had no time to finish this one… =/

The secret voice on the videos is from Filipe Calegario (calega). Note the well mixed brazilian+german english ascent. ;)

Hopefully we can see some these as one of the 5 Nokia PUSH winners!

And a personal thanks to all wouwlabs people and patifa for all
the efforts they put on this. You guys rock! ;)

music

2009-10-17T20:47:00+00:00

Here is the MP3 file (or MIDI file) of my First Random Song .

I used JFugue to deal with MIDI stuff and implemented the code that generates chords sequence, notes duration and basic melody following some simple rules (e.g. the notes are from the C Major pentatonic scale and the chords are the simple: C F G Am).

Despite the simplicity, I really enjoyed the result. =D Well, I’ll give more details about the implementation later.

This is the first test of a project in which I’m studying new ways of composing music using multitouch tabletop interfaces. I’ll also talk about it later.

PS: the picture above I painted using ArtRage 2 Started Edition, it’s quite nice to play around with the tablet using this software. I don’t like to post anything without a picture… =)

Circuit Bending no Continuum

2009-10-16T19:30:29+00:00

A primeira edição do Continuum está acontecendo no Recife desde semana passada e conta com um bocado de atividades super empolgantes para qualquer um interessado na combinação de Arte e Tecnologia, seja para bater um bom papo, colocar a mão na massa ou curtir os mais variados experimentos audio-táctil-visuais.
O feriado em Recife talvez tenha afastado o público de um interessantíssimo woskshop realizado por Cristiano Rosa (panetone) tratando de Circuit Bending. Apesar da lotada pré-inscrição, poucas pessoas se apresentaram para trocar idéias e aprender um pouco sobre como dar voz à nossa aparentemente inutilizável sucata eletrônica. As coisas começaram com Cristiano mostrando um pouco do que tinha feito, como um hacking para um Furby (um gremlin-like robot toy) e alguns módulos que atuavam como filtros para um sequenciador, tudo feito de uma maneira muito simples. Ao contrário da maioria dos tutoriais sobre “como montar sua própria buginganga-eletrônico”, ninguém precisou ter um background sólido sobre componentes eletrônicos para começar a desmontar os toca-fitas, joysticks e teclados e criar os efeitos mais inesperados.

É incrível a diversidade do potencial de criação que temos à mão se não esperarmos pela prateleiras. Sem sombra de dúvida vale à pena dar uma olhada no livro do Reed Ghazala, pioneiro e difusor no Circuit Bending para começar com umas boas (des)montagens.

p.s: o próprio panetone irá se apresentar no Continuum, na Torre Malakoff, no próximo Domingo, tocando nos instrumentos que ele criou.

Imagem1

2009-10-05T14:18:47+00:00

algum tempo atrás, enquanto falava aqui mesmo sobre o LIBRES, comentei sobre a grande lacuna existente no Recife de eventos com a temática de Arte & Tencologia e sobre um suposto início de mudança de cenário. nada melhor pra confirmar essa hipótese do que a chegada do Continuum – o I Fesival de Arte e Tecnologia do Recife. segundo o próprio site do evento:

Continuum é mais do que um evento que reúne arte e tecnologia. É um festival que aproxima informática, cultura, pesquisa livre e diversas áreas de conhecimento (Cinema, Música, Literatura, Design, Moda, Informação) em uma abordagem multidisciplinar. Há muito o que ser descoberto, há muito o que ser revelado e há muito o que aprender neste mundo atual tão carregado de informações e dados intercambiáveis.

o evento, que acontece entre os dias 8 e 18 de Outubro, vai rolar na Torre Malakoff e no Centro Cultural dos Correios, contando com workshops, palestras, além de várias obras em exposição. pra quem tiver pela cidade na época, fica aqui a dica!

maiores informações podem ser encontradas no site.
ps 1: a correria anda grande… muita produção ultimamente!

ps 2: pra quem não sabe, meu trabalho com o mtPd foi premiado no Itaú Rumos Arte Cibernética, carteira de pesquisa acadêmica! tô muito feliz! espero em breve falar um pouco mais sobre o projeto, quem sabe até com alguns vídeos!

Tagged: eventos

The Type Tester

2009-09-27T17:09:33+00:00

Hi, everyone!

It’s been a while since I don’t post anything, and that’s mostly because of the university – it is giving me a lot of annoying headaches. When you get to the end of the course, there are somethings that you just can’t handle anymore, like exams. And also, now I’m working too (yey! : ), so it gives me too little time to do stuff that I like.

Every time when I have spare time, though, I try to work on the development of the widget explorer and it’s integration with plasma – actually the last time I did that was last weekend… I did some useful commits, but now I’m kind of in the middle of too many exams and projects at university, so I can’t do much. Sorry . Aaron is the guy who’s helping me out with this, so, Aaron, thanks, you’re the man (ps: please pong me at irc someday)!

Anyways! It’s been a while since I’m working on a plasmoid idea and I think it’s time to share it with you.

It’s a pretty simple plasmoid that I’ve been developing since the GSoC period (well, I didn’t have much time to work with it, since I had a GSoC project to accomplish), and now I’m getting it finally done. I call it Type Tester.

I don’t know about you, but everytime I have to do a task that includes choosing a font type to some certain text, I takes me forever to do it! I copy and paste the phrase I want to change the font a few times and start changing its fonts. Looks like a quick task, but it’s not at all. Then I start comparing then, and deleting the ones I don’t really like, until there’s only one left! Stupid?!

I always thought about building a solution to this problem I have, and now I finally found the perfect solution: a plasmoid to test the fonts! It’s quick, simple, and harmless

The idea is simply to type a text, choose a font size and build a list with this text written with each of the fonts of the system. Here is a screenshot:

(ps: ignore those fonts I have – I installed them when I was little (yes, like 11 years old) and I didn’t know what I was doing back then : )

(ps2: I know, I still have to handle the resizes stuff!)

When clicking in the “-” sign, that font will dissapear of the list. For the future I want to be able to:

do actions in batch, like erasing the selected ones, or erasing the not-selected ones
change the text foreground and background
change the number of fonts that appears on the list per page (tab)
accept a font point size number input
name the tabs with the first letter of the fonts names on that list (A / B / B..C / C / D..E).

If you like the idea and want to help me with new ideas, please do it

cheers!

Tokamak3 and Widget Explorer evolution

2009-09-12T22:58:05+00:00

Hello there!
It’s been a couple of days that i’ve been willing to write this post, but I hadn’t find enough time yet. Since I missed a lot of classes when I went to Tokamak3 (which, btw, was completely, totally worth it), my life at university has been crazy since I arrived here.

Well, to begin with, Tokamak3 was great – one of the most significant/exciting/pleasant/enriching experiences I’ve had. It is impressive how few people can get together and be so, so, so productive in so short time, and, at the same time, get along so well, having fun, laughing and being friends. Yes, it’s beatiful to see the friendship going on inside this community. They are all so cool!
Every single one that was there impressed me on your own specific way – each one with your peculiar personality, but all of you with your great technical skills (the same to you, Mario, with your cooking technical skills . That was quite inspiring to me, the fresh new member. Yes. That was VERY inspiring… …
Damn, I miss Tokamak!

Ok, but now for the news: after Tokamak3, the new widgets explorer is now merged with trunk! And now it is not the New Widgets Explorer anymore, is THE Widgets Explorer!

On the first day at Tokamak3, we all did a presentation about what we were going to work during the next days, and after that we wrote on postits the tasks we were going to accomplish during Tokamak. That was a pretty interesting idea, cause this way we could keep the track of our work, knowing what we have done and what’s left to be done.

Before going to Tokamak and writing my tasks, I compiled all the hints, criticism and ideas I’d received with my last post and discussed them with the people at the chalet. Some were incorporated, some were saved, some were not.

The first immediate idea was to solve a problem one pointed out: it isn’t so obvious that I have to drag and drop the applet to add it. Ok, I agree. So the solution was to change the mouse cursor to a little opened hand when hovering the icon, and changing it to a little closed hand when clicking on the icon. Nice and easy, huh? Task solved!

Another task was to highlight in the list, somehow, the already running applets. It’s nice to know what are the applets that you have running while browsing throught the list. The old widgets explorer had this functionality, but the indicator of running applets was “-” sign, which was also a button to delete those applets – this was pointed out to be an usability problem. The solution to that was to paint a “tick” blue sign on the left-bottom corner of the running applets icons. This “tick” sign is also the icon to the “running” categories, so the user knows what that tick means. Task solved!

Another set of tasks was to fix a few layout problems – I called it the crazy layouts, but now they are under control.
Task solved!

We also had some discussion about the problem of removing applets through the widgets explorer. How to do that? If we had some kind of remove button for each icon of running applet in the list, then we would go back to the problem of the old widget explorer: the remove button simply removes ALL of the widgets, which is not nice and not expected. And if we had a remove something for each single plasmoid of every desktop, than we would maybe get to an usability issue. The solution, then, was to not implement this functionality inside the widget explorer. To remove a plasmoid you can simply remove it through the plasmoid handle.
Task solved!

Other interesting task was to find a way of showing the brand new kind of widgets: the remote ones (Rob’s summer project). The idea is simple: show the list of widgets that are available on the network. We find one solution that was: create a “remote widgets” category. But there are still a few questions to be answered: do I want to filter the remote widgets with categories too? If so, remote widgets can’t be a category itself. Do I wan’t the remote widgets to be showed among the local ones? Do we want to publish/unpublish widgets on the network through the widgets explorer? Ok, this question has been answered: no, we don’t. But now all we need to do is to answer these other questions and implement it. Task partially solved!

The most polemic non-technical task was to figure out a better way of displaying the vertical version of the widget explorer. There is something not right with the list of categories… And no, we DON’T want a combobox! Aaron came with the idea of a “kind of” combobox, that expands with mouse hover, and doesn’t hide when choosing a category (cause a combobox hides after selection), so we can filter quickly more than once in a row. How about that? Actually, after writing it down, I found it a very nice idea! I can already visulize it, it’s a new better kind of combobox that may actually work. But I open it for suggestions. Any?! Ok, this task was not solved.

I think we’ve reached the last task: merge the widget explorer with plasma! This is the most important task – put the widget explorer inside the panel controller. Yes, that was where we were going to begin the merge – clicking add widgets on the panel controller, the widgets explorer would appear. Aaron helped me out with the beggining of the task, and then the most time of Tokamak was spent to make this merge work. I was facing a lot of layout, dinamic resizing problems, but the task to put the explorer inside the panel controller was solved! You can build trunk and test it But there are still quite a few tasks to be done to make it usable and prettier. And we still have to figure out and make it work when right-clicking on desktop and when calling the up-right corner button to reach “add widgets”.

There is a list of remaining tasks to be accomplished and i’m going to quickly list them.
Feel free to add some to this list

Put a “close” and “back” button on the widget explorer
Make the panel controller wider when clicking on “add widgets” (so the icons won’t appear too small)
Maybe make it possible to the user to resize the widget explorer
List the remote widgets
Make it possible to uninstall an applet through the widget explorer
Figure out the whole vertical list
Make it work when right-clicking on desktop and when calling the up-right corner button to reach “add widgets”
The arrows button looks disabled, I know (with air, at least). Need to figure that out.
How about making the panel controller a little more translucent?
Yes, I know the name of the applet is clipped, I’mma fix it
put an info button inside the plasmoids handle to show more info about the plasmoid (author, version, etc), since we removed it from the widget explorer.

Well, I guess that’s it. There’s a lot going on and I’m very happy with the work we are doing with the widget explorer, and the progress we made in Tokamak! Let’s keep working and, please, if everyone wants to help, do it (there is a lot helping already, I just want to make it clear . With ideas, hints, coding, code hints, and so on.
We have to make it work to 4.4

Oh! I forgot to mention: also during Tokamak 3 I received a very good news! I was accepted as a Speaker on Latinoware 2009 (http://www.latinoware.org) ! My presentation there will be about my GSoC experience, aiming the ways of gathering/working/developing UX in such a “spread” way as Open Source Projects are. It was so nice to see my name on the speakers list (http://www.latinoware.org/node/40) ! Hope to make it well and to see some of you there!
(what about a quick kde sprint while we are there?!

And I’d like to finish this post by thanking KDE e.V. and the whole KDE community for every single opportunity on the past few months. You guys _really_ rock!

See ya!

gallery

2009-09-05T00:04:41+00:00

releitura da obra de Prokudin: "Seurat vai à Rússia"

Seurat

Georges-Pierre Seurat (Paris, 2 de dezembro de 1859 – Gravelines, 29 de março de 1891) foi um pintor francês e pioneiro do movimento pontilhista, também chamado divisionismo. (..) Em 1880, torna-se Mestre, inspirado pela obra de Michel Eugène Chevreul: A lei do contraste simultâneo das cores(1839). (…) A técnica do pontilhismo utilizada por Seurat deu origem ao neo-impressionismo e foi extensivamente utilizada na arte do século XX. Pode-se dizer que a teoria pontilhista foi precursora da televisão e da imagem digital. (Fonte: Wikipedia)

conheci o trabalho de Seurat por esses dias, através de um livro que ando estudando sobre sintaxe visual, que me trouxe uma atenção bem especial aos pontos. nunca tinha parado pra pensar muito sobre eles… explorar seu potencial foi um dos objetivos deste exercício/projeto!

releitura da obra de Prokudin: "Seurat vai à Rússia"

Prokudin-Gorskii

Em 1905, Prokudin-Gorskii concebeu o grande projecto de documentar, com fotografias a cores, a enorme diversidade de história, cultura e avanços do Império russo, para ser utilizado nas escolas do Império. Com o apoio do Czar Nicolau II, percorreu o império entre 1909 e 1915, documentando-o com imagens e dando a conhecer a magnitude deste aos seus habitantes. Do resultado do seu trabalho surgiram as primeiras patentes de filmes positivos a cores e de projecção de filmes com movimento. (Fonte: Wikipedia)

imagens coloridas da primeira década do século passado. não tenho palavras pra descrever o quanto elas me sensibilizaram (tomei conhecimento pelo twitter, via @glerm)… era como olhar direto num túnel do tempo! muito foda!

releitura da obra de Prokudin: "Seurat vai à Rússia"

Sauret vai à Rússia…

inspirado nesses dois elementos (e numa sexta feita livre de maiores obrigações) e baseado no Pointillism, de Daniel Shiffman (de fato, é um exemplo disponível dentro do próprio processing), a aplicação reconstrói as fotografias de Prokudin sob uma nova perspectiva, através do uso ostensivo de pontos de diversas dimensões e cores, num o processo semiautomático (o usuário pode interferir até um certo limite no processo de construção das imagens). apesar de simples, achei o resultado final bem interessante – digno de um post aqui no blog!

outras imagens podem ser conferidas logo abaixo. espero que gostem!

maiores informações

[1] sobre Georges Seurat

[2] imagens originais de Prokudin-Gorskii

[3] sobre o processing

Tagged: processing, projects

eurotrip

2009-09-01T20:41:24+00:00

22 days. 5 Brazilians. Tons of luggage. 1 cajon. 1 ukulele. 1 Fiat Punto. 3878 kilometers. 3 countries. 1 city-state. 1 principality. More than 15 cities. Lots of people we met. Lots of things we learnt. Thousands of photos. No more words.

View Google Maps

Thank you, GSoC. Now let’s move on!

2009-08-25T04:01:25+00:00

Hello, everyone!
It’s been a while since I don’t post anything, and very sorry for that. You soon will find out my possible reasons (or maybe some excuses).

This week the GSoC season ended and I am here with my first of the last posts about the GSoC

First of all, I’d like to thanks everyone that made it possible and made it beautifully – including gsoc team, kde/plasma community, the qt labs america guys and, of course, Ivan, my dear, patient mentor. Thanks you all.
This was, and I really mean it, one of the most significant experiences I had – I managed to overrun myself in several aspects, since the first day I joined the kde community. Besides all the technical learning, I managed to keep a nice long-distance relationship with lots of people, and find a perfect virtual environment to work. It’s being just great.

Well, as you must know, I am working with the New Widgets Explorer, and this particular project also brought me lots of learning.

One of the most difficult tasks I had to deal with was to conceive the new UI & funcionalities of the project, and that started even before coding time began.

The Widget’s Explorer is an everyday tool for plasma users, so almost everyone that uses it has a strong opinion about it – likes, dislikes, how things should or shouldn’t be. It is obviously very hard to please everyone, since we’re all so different. I had to be prepared to hear lots of criticizes!

My first attempt to do the right thing was to arrange a virtual focus group session with plasma users – I tried my best to gather different kinds of users: developers, designers and purely users. There were 12 people in the room, and the results were great! I could analyze a huge material and came out with some nice conclusions (you can see it in previous post).

My second step was to try to express the conclusions into mockups. I drew several ones, and finally published 2 options of mockups, which you can also see in previous post (the last one). The response to those was great. I heard a lot of criticizes, got a lot of help, lots of private e-mails and blog comments with suggestions. The final choice was to use one of the mockups and embrace some helpful criticizes.

So we were done for now with the mocks. Coding time begun! Yey!

Everything was going great and I was able to create a functional mockup-like (not so much alike) version of the project. I was very excited with that, mostly because of the huge learning.

But, as we know, it is hard to please everyone – the whole idea of the project (the UI aspect) was questioned and we decided to take it to another path.
First of all, this is a screeshot I took before changing paths (not the most up-to-date one, I had more stuff done – but this was a screenshot I luckly found here ) :

The new idea, though, was to make it a long and linear list, horizontally attached to the panel – basically. So, after developing a great part of the project, I had to stop and make some other mocks.

This is one that could sketch, with the help of some community friends – the horizontal and vertical versions (yeah, vertical ’cause we needed to think about the case when the panel is on the right/left sides of the screen):

Some work was lost, and lots of hours of work were lost – more hours than code lines (those I could reuse), ’cause, as a begginer, I didn’t have the experience to code so many lines in few hours. But this whole re-working thing made me learn much more

The new idea/mockup for the project didn’t came out to public due a decision I made: I had to be able finish GSoC in time and I couldn’t be doing any back and forth work all the time (designing – coding – re-designing – re-coding) and I knew that there would be a long time of discussion over it if I did. I find this kind of discussion very productive, but I really had to finish GSoC. And, yey!, I did it!

I finally finished what we agreed to do after the whole scenario changing, and these are the screenshots of the final result
(screencasts later:)) :

There are both vertical and horizontal versions implemented. The categories are always visible, and tooltips holds more information about the applet, and also moves with animation like the plasma tooltips. The list moves with the arrows buttons and also with mouse wheel (faster with the mouse wheel). These tooltips were great idea – it can be a place for lots of stuff, like the screeshot of the applet, the number of running applets and its position in desktop…

Hopefully you’ll like it

But I am actually here to open up a space for every of you opine over it. Now that GSoC is over, I don’t mind doing back-and-forth job (for as much as it is better for the application ).

Of course there is still a lot of work to do, and I’m thrilled to raise our new child together with you!
The day after tomorrow we’re heading to Tokamak and I’m absolutely sure that the New Widgets Explorer will mature much more with those terrific guys (and girl )

Well, later I’ll have another post with screescasts and some other details

jeraman

2009-08-19T14:55:57+00:00

no último post falei sobre um projeto em que eu e o pessoal da Code Media estamos desenvolvendo em parceria. pois bem, é justamente ele o tema deste post: o “vi”!

o “vi” é como chamamos um projeto de uma vitrine interativa. vitrines interativas são basicamente um sistema interativo em forma de vitrine que interage em tempo real com as pessoas que estão passando por perto desta, de modo controlar a aplicação central – que por sua vez ficará visível na vitrine (no lugar onde tradicionalmente ficariam os produtos de uma loja). nossa inspiração inicial foi um trabalho bem interessante que realizaram numa apple store, lá em Berlim, cujo vídeo pode ser conferido logo abaixo:

o esquema técnico desse primeiro release se baseia numa webcam, que envia a imagem para um software que a processa e detecta a posição exata da pessoa. o responsável por tal processamento é o Community Core Vision (CCV), software amplamente usado no contexto de aplicações multitoque, desenvolvido por Seth Sandler, do NUIGroup. apesar de não ter sido criado para esse objetivo, o software vem quebrando bem o galho… a partir daí essa posição é então enviada via sockets para a aplicação Flash, que trata de dar o retorno visual da aplicação.

o resultado final do release pode ser conferido no víde lá em cima! ainda haverá outros releases, com uma série de incrementos… em breve mais notícias!

links interessantes

[1] site do Community Core Vision
[2] site da Code Media
[3] site de Seth Sandler

Tagged: flash, of, projects, vitrine interativa

2009-08-17-132006

2009-08-17T16:35:06+00:00

webcams desempenham atualmente um papel extremamente importante em aplicações interativas, sendo utilizada como principal meio de entrada de informação. isso nos faz ter idéia da importância decisiva que a escolha de uma webcam adequada – tanto do ponto de vista de desempenho, quanto (para os lisos como eu…) no quesito preço – configura para o sucesso desse tipo de trabalho. levando tais parâmetros em conta, torna-se facilmente explicável porque a PlayStation 3 Eye, da Sony, vem se tornando a webcam favorita dos cientistas loucos de plantão…

criada inicialmente como um acessório para o PlayStation, com um preço médio de $40 (muito barato quando comparado com câmeras de qualidade semelhante disponíveis no mercado), o equipamento logo chamou a atenção da comunidade opensource (créditos à galera do NUIGroup!), principalmente por ter sido construída para lidar com aplicações que envolvessem visão computacional especificamente… foi o suficiente para motivar a galera a desenvolver drivers específicos nas 3 plataformas, os quais nos permitem hoje usá-la como uma outra webcam qualquer, para nossa alegria!

foi justamente essas qualidades que motivou o pessoal da Code Media a adquirir uma, para um projeto bem interessante que estamos desenvolvendo em parceria (em breve espero dar maiores informações!).

no nosso caso especificamente, foi necessário que adaptássemos a PS3Eye para que ela passasse a captar apenas raios infravermelhos, desprezando a luz visível. esse passo essencial em alguns tipos de projeto (tal como o da mesa multitoque). o processo de modificação foi fácil e rápido (no máximo uma hora de duração) principalmente graças a um vídeo feito por Peau Productions, do NUIGroup, que pode ser conferido logo abaixo:

o último passo foi instalar a câmera no linux, processo também bastante tranquilo e pode ser conferido em detalhes aqui (antes de continuar, confira se vc possui o source do kernel devidamente instalado na sua máquina: tive uma grande dor de cabeça por conta dessa besteira… valeu jeez pela ajuda!). fizemos também os testes em outros sistemas operacionais e ela mostrou um bom desempenho em todos, sem exceção!

resultado final pode ser conferido no screenshot abaixo: a PS3Eye modificada rodando a 60 fps no Ubuntu!

links
[1] review interessante da ps3eye
[2] tutorial: instalando a ps3eye no linux

Tagged: ps3eye

libres 2009

2009-08-06T13:31:30+00:00

que Recife é uma cidade com uma grande enfervescência artístico-cultural, todo mundo já sabe. ao mesmo tempo, a cidade possui um grande polo tecnológico (Porto Digital, Centro de Informática…), sendo uma referência nacional na área. tais características quando reunidas parecem favorecer o surgimento de uma nova corrente, que visa justamente agregar esses dois elementos num novo meio de expressão, o que pode ser facilmente percebido pela quantidade crescente de pessoas que passam a se interessar pelo tema…

dado esse cenário, era de se admirar a carência de eventos sobre arte & tecnologia numa cidade tão louca e propícia a esse tipo de coisa como Recife. pois bem… para nossa felicidade, tal carência começa a ser deixada de lado com o LIBRES.

o LIBRES é um evento multimídia de arte e tecnologia que pretende desmitificar o conhecimento na área de arte e tecnologia, disseminar este saber, estimular sua apropriação e pesquisa, bem como divulgar seus frutos. para isso, se pretende intercambiar e divulgar os trabalhos e possibilidades artísticas e técnicas de multimídia interativa criadas com ferramentas baseadas em uma cultura livre para apropriações, recriações, mixagens e incrementações. a proposta abrange a troca de conhecimentos entre pesquisadores, desenvolvedores, artistas e público em geral, no intuito de estimular o uso destas plataformas no exercício de suas criatividades; assim como fomentar o aprendizado de processos tecnológicos através da interação entre obra tecnológica e público.

o evento vai do dia 6 até o dia 8 de Agosto, lá na Torre Malakoff. toda a programação (que por sinal, além de contar com um pessoal de fora, tratou muito sabiamente de convidar uma boa parcela de artistas locais) pode ser conferida no site do evento. pros interessados, uma ótima oportunidade de imergir um pouco mais no tema!

Tagged: eventos

42. Encontro Livre - dia 3

2009-07-30T13:53:57+00:00

nos dias 6, 7 e 8 desse mês, rolou na Livraria Cultura, no Recife, o I Encontro Livre – disseminando cultura e conhecimento. o evento foi muito interessante, contando com diversas atividades como palestras (teóricas e técnicas), debates, filmes, além de contar com um bom público, bem participante… em suma, um sucesso! fotos e os slides das palestras podem ser obtidos no site do evento!

no 3º dia, entre outras palestras técnicas, nós do wouwlabs (representado por mim) tivemos o prazer de dar uma palestra sobre tecnologias multitoque: a “Come and Touch it! Open Multitouch“.

nessa apresentação teórica, falei sobre o funcionamento básico de uma superfície multitoque, das tecnologias e comunidades envolvidas, além dos problema encontrados atualmente. ao fim, tentei fazer uma demonstração prática que – infelizmente – acabou não funcionando muito bem devido ao curto espaço de tempo que tivemos para montar todo aparato… uma pena!

a palestra foi bem legal e o pessoal parece ter gostado bastante! tomara que ela tenha servido um pouco para desmistificar um pouco tecnologias multitoque em geral e aliciar novas mentes inocentes (muuuuuarrrrrr!) para trabalhar com elas, ajudando a construir (quem sabe?) uma comunidade local!

um show a parte ficou por conta da galera do Coletivo Make Install, representados por Ricardo Brazileiro e Felipe Machado. eles mostraram um trabalho chamado Emotional Kernel Panic: um modulo do Pd que gera um retorno audiovisual baseado na atividade do kernel do Linux… coisa fina!

perdeu a palestra? bem… de qualquer forma, vc ainda pode conferir os slides logo abaixo!

Tagged: lectures, multitouch

macuntu

2009-07-06T23:25:37+00:00

finalmente consegui colocar a maria creuza pra rodar com dual boot: usando o osx leopard e o ubuntu studio 9.04.

apesar do processo todo (incluindo configurações de wireless, bluetooth, teclado, atualizações e coisas do gênero) ter durado 2 dias (grande parte dessa demora motivada por minha grande inexperiência no assunto) pode-se dizer que o saldo final foi positivo: o sistema parece estar bem estável, sem maiores incompatibilidades.

para aqueles que gostariam de fazer o mesmo, fica aqui alguns links bastante úteis!

[1] informações gerais sobre o ubuntu em macbooks intel.
[2] how-to detalhado da instalação (aqui e aqui)
[3] caso tudo dê errado, confira aqui como formatar e particionar o HD

Tagged: linux

Computational Photography

2009-07-05T13:52:41+00:00

This is the result of a Computational Photography assignment I finished a couple months ago.

http://www.filipecalegario.com/comppho/3assignment/

DSCF8815

2009-06-30T07:00:28+00:00

3 months in Munich. Lots of things learned. Lots of interesting projects. Understanding a little more German! =D

jeraman

2009-06-29T22:05:29+00:00

depois de quase cinco anos de sofrimento e quatro meses de trabalho intenso no projeto, finalmente é chegada a hora: apresento nesta próxima quarta feita, às 10h da manhã, lá no Centro de Informática, o meu trabalho final de conclusão de curso, vulgo monografia.

o resultado do trabalho, o mtPd, uma interface multitoque para o Pure Data, pode ser conferido em detalhes através da monografia (que pode ser baixada aqui), ou ainda através dos slides da apresentação final (que pode ser vista aqui).

em breve espero fazer um vídeo da ferramenta, mostrando como ela funciona em detalhes, além de abrir todo seu código fonte. uma parte já foi liberada e pode ser conferida aqui mesmo, neste blogue.

resumo
Surgidas na década de 50, a partir do Music I de Max Mathews, as ferramentas para processamento de áudio desempenham atualmente um papel importante no cenário musical, utilizados ostensivamente por bandas e DJs, o que reflete as centenas de ferramentas para processamento de áudio em tempo real disponíveis no mercado.

Neste contexto insere-se este trabalho: ele apresenta o mtPd: uma ferramenta de processamento de áudio em tempo real baseada em um ambiente multitoque, concebida de maneira a ser expressiva, intuitiva e colaborativa para lidar com performances artísticas ao vivo.

maiores informações:
[1] link para a monografia
[2] link para a apresentação final
[3] sobre o pyPd
[4] sobre o Pure Data (Pd)

Tagged: pd, projects, python

filipecalegario

2009-06-28T21:56:39+00:00

Last trip to Dublin. Heineken, Guiness and with “saudades de” Faringes da Paixão.

filipecalegario

2009-06-27T15:43:42+00:00

http://twitter.com/filipecalegario

for those who don’t know the idea behind it:

filipecalegario

2009-06-11T15:46:51+00:00

I am posting those pictures to keep them here and serve as inspiration for my future projects in Information Visualization.

For further information:

http://marx-blog.de/2009/05/social-media-infografiken/

http://www.visions.com.ua/70-prekrasnyx-primerov-infografiki/

First mockups!

2009-05-21T22:17:59+00:00

Well, here I am again!

Coding time is coming and I’m very excited about it. I’ve learned a lot this month and things are coming out great for me – I’m really liking the community, people are very, very nice, and also I have a very cool and responsible mentor

But today it’s about mockups!

In my tentative timeline I said I was going to do some mockups drawing and deliver it to the community approval, so we could built together the actual, real and beautiful mockup. So it’s time, right on time!

First of all, we came out with a list of functionalities the widgets explorer will have:

List applets
Filter by category
Filter by most used (support of usage statistics)
Drag and drop to add widget
Support of emblems
List kde-look widgets, download source widgets
Install kde-look widgets
Install from directory

These are the required functionalities. Of course desirable funcs can appear and be a part of it (some already appears in mockups). The great thing is that we were able to remove functionalities that weren’t used very much and then a more simple UI could be built.

I managed to draw 2 ideias to the widget explorer, where I tried to put all the functionalities, UI, usability concepts we extracted from focus group, and also others I studied. Some of these concepts were: no scrolling bar – instead we use navigation buttons to reveal pages; more information about widgets when navigating; categories always on, not hidden in combobox.

These are the drawings:

I talked to pinheiro (great kde designer) about the mocks, and he really enjoyed it, particulary the second one. He said we could go for it, but first we have to ask for the community approval. Once is decided, pinheiro said he could do the actual, real, beautiful mockup, like no one can do

When drawing the mocks, I represented the widgets as a little ball, as you can see. That would represent the widgets icons, actually. This icon issue was largely discussed at focus group, cause there is too many icons to be made, and too few designers to accomplish that. The truth be told: these icons are too cute to be discarted. So I thought: what if these icons were all little translucid plasma-like balls, with some content inside? I think the work of doing icons would be much easier and more people would be capable of doing it, since the content would be much simpler… I don’t know. It was just an ideia… Pinheiro agreed with me, though! “I can already imagine how it would like with little balls ”

cute

well, I’m looking forward to hear your feedback!

el-448x209

2009-05-12T18:10:43+00:00

os queridos amigos ex-agorianos [celia] e seucolega estão organizando o I Encontro Livre – disseminando cultura e conhecimento. o evento, que tem o intuito de divulgar o uso, o desenvolvimento e os benefícios do Software Livre para os diversos segmentos da sociedade e promete ser bastante interessante, será realizado na Livraria Cultura, em Recife, durante os dias 06, 07 e 08 de Julho de 2009, entre as 17h e as 21h. se vc estiver pela cidade na época, vale a pena dar uma conferida!

sobre o evento

Desde meados dos anos 70, época em que era comum compartilhar código-fonte, o Software Livre está cada vez mais presente no cotidiano de residências, ONGs, empresas e governos graças às vantagens de ordem econômica, tecnológica e ambiental. Seu crescimento é tanto que, no ano de 2001, a Prefeitura do Recife regulamentou uma lei de incentivo ao uso dos programas de código aberto nos sistemas e equipamentos de informática de todas as secretarias e empresas da administração indireta. Desde então, várias prefeituras em todo o Brasil estão adotando o Linux como estratégia econômica para reduzir gastos, ampliar conhecimentos, estimular novos aprendizados e contribuir para o desenvolvimento de tecnologias da informação.

O I Encontro Livre será realizado nas dependências da Livraria Cultura Recife, durante os dias 06, 07 e 08 de julho de 2009, entre as 17h e as 21h, com o intuito de divulgar o uso, o desenvolvimento e os benefícios do Software Livre para os diversos segmentos da sociedade. A programação, destinada a pessoas interessadas em conhecer ou consolidar conhecimentos, estudantes, acadêmicos e profissionais da área de Tecnologia da Informação, abordará as áreas de desenvolvimento e filosofia.

para maiores informações, visite a página do I Encontro Livre!

Tagged: eventos

Imagem1

2009-05-08T03:11:50+00:00

today i have a great announcement: pyPd! this is the project that has been keeping me busy for a while…

pyPd is a multi-platform and open-source python library that allows people to use Pure Data just like an API. in other words, it is an abstraction to Pd, that can be used for doing everything that Pd can do, but without opening Pd interface in a direct way.

sample of Pd patch. with pypd, you can now build one of this using python!

it was developed using a technique known as “dynamic patching” (already cited in this blog), which is based on dynamic creation of Pd objects realized through sockets. this means, for example, that we can create a [osc~ 440] object sending to Pd a specific message through sockets.

so what basically the library does is to receive the commands given by users, transforming them into the equivalent message. these messages are sent to Pd, that processes them like they were a patch running inside Pd (a virtual patch), generating results. after that, these results are reloaded again into the python program through simple file reading, what allows it to be always updated according the “virtual patch”.

it’s important to notice that no gui is shown during this process. Pd runs in -nogui mode, so all interaction happens through messages by sockets and file reading!

this first version is quite simple. it includes just:

object creation, edition and remotion (object, number, comments, symbols and messages)
connection management (connect and disconnect objects).

for further reading, the project (including the source) is available at:

http://code.google.com/p/pypd/

get started reading the tutorial! it’s not hard indeed!

ps: pyPd… till now i don’t have a better name… suggestions?

further informations

[1] about Pure Data
[2] pyPd project on google code
[3] getting started with pyPd
[4] about dynamic patching

——

i’ve got a lot of news… here comes some of them:

yeah… i was not selected to participate of gsoc this year, but that’s ok… maybe next year? the entire list of projects selected by NUIGroup can be checked here. congrats and good luck for all students!
but… anniec, member of wouwlabs, had her proposal accepted! this is just awesome! congrats anniec!!! we all are with you!!!
well… i’m fired! crisis to blame… now i have more time to lead with my thesis… =P
a new tbeta (aka CCV -Community Core Vision) is out! go and check it!
it’s been a long time without posts and i’m sorry for that. i spent the last three weeks taking care of my life and thinking about what the hell i’m gonna do after graduated… so i didn’t give too much attention to the blog…

Tagged: pd, projects, python

Focus Group Results – All about the widgets

2009-05-01T19:06:09+00:00

Hey, you all. It’s been a week since our focus group session happened and I spend this week doing some analysis over it. I gotta tell you – it was a very productive meeting, with some very good ideas and discussions that should lead us to very good results! For the ones who asked, mockups are coming soon (don’t worry – designers will help me with that , but not now. I’m trying to follow my timeline strictly

Before I start talking about the results, I’d like to thanks to all of the 12 people who participated of the session:

MoRpHeUz, jeez__, guiocavalcanti, pinheiro, notmart, ivan_cukic, aseigo, cmarcelo, anselmosm, pinda, annieC, seele

Thanks for helping this matter!

Well, first of all, me and pinda had a script to follow during the session, with topics to guide us. Since those topics eventually changed courses along with the meeting, I’m going to write down the actual topics found in the discussion:

How often do you use the widgets explorer?
Do you find it useful? Why?
What do you feel about it?
Which functionalities do you use?
When you’re searching for widgets, what do you do?
What do you feel about this search mechanism and the way the widgets are listed?
Which functionalities bothers you?
What usability problems do you feel about the current widget explorer?
And what about functionalities issues?
There are a lot of ways remote widgets can possibly be displayed in the add new widgets dialog. What way do you think is most comfortable to use?
I understand this question concerns a feature that doesn’t exist yet, but just try imagine having that possibility. Where would you expect them to be? Like in a seperate tab/category? mixed with other widgets with a different emblem? structured by computer that shares them?
But what do you think? Consider each other machines like a seperate categories? Have one category
‘plasmoids near you’? Have a tree-view like thingy that structures widgets under the computer that shares them?
Terminology for this widgets-over-network feature: what do you think a good name
would be for accessing a widget this way?
Take a look at these mockups and talk about it:
http://www.kde-look.org/content/show.php/Plasma+-+Add+Widgets+Mockup?content=101423
http://blusrcu.ba/nookie/?p=22

The topics 10 to 13 where pinda’s topics. We made analysis of those to, but pinda is the one who’s gonna explain that to you.

Below are 2 mindmaps of the focusgroup results. I put it in short topics for a quick understanding of the session (for those with no courage to read everything )

This one sums up the discussion about the current widgets explorer and mockups

This one were the ideas that appeared when discussing the current widgets explorer and mockups

Ok, so I’m going to analyze topic by topic.

1 – How often do you use the widgets explorer?

2 – Do you find it useful? Why?

That would be pretty obvious to say that the widgets explorer is used only when there’s the need to add widgets. It became clear, thought, that’s not an everyday/very often used tool – but it’s still very important and useful. The main point is that the tool is used in a low frequency and for supposed quick tasks.

3 – What do you feel about it?

As known, the current widgets explorer doesn’t satisfy much – seems that it doesn’t answer to it’s purpose. Incredibly, even being a not-very-often-used tool, it can bothers users pretty much – everyone at the focus group session. So much it makes one wanna kill it.

<@annieC> and what do you feel about it?

<+pinheiro> the usuall “kill, kill, die, die

Widgets explorer seems to be poor to the users – it has a lack of plasmoids info [jeez__], no organization and very annoying ugliness [pinheiro]. A bit flat list doesn’t please anymore [notmart].

4 – Which functionalities do you use?

This topic was about raising the most used functionalities of the tool and keeping in mind that those should be treated more carefully.

The first functionality remembered at the session was the drag and drop. The “add widget” functionality is discarded with the possibility of dragging and dropping the plasmoid. [notmart] exposes that he uses drag and drop specially because it puts the widget exactly where he wants. [jeez__] said, though, that he took sometime to realize he could dnd the plasmoid – an usability issue observed that maybe can tell us something.

There was a conflict about the “favorites” functionality. Some says it doesn’t make much sense, others don’t use it much, others don’t use it at all. [pinheiro_] brought a strong opinion when said that he tend to prefer “most used” categories rather than favorites, since the “most used” list is made automatically.

5 – When you’re searching for widgets, what do you do?

With this topic, things can get a little confused. [ivan_cukic] said that he searched by “favorites” – functionality barely used be others – “most used” and by typing the name. He said he doesn’t really uses categories.
It became clear that most of them uses the search field, either to search for functionality or to search by name.
The categories are ignored by most of them – some said because it’s hidden under a combo box. [cmarcelo] thinks, though, that the use of categories could be nice to discover new widgets, even though he doesn’t use this search mechanism.

6 – What do you feel about this search mechanism and the way the widgets are listed?
7 – Which functionalities bothers you?

There were a few issues that can be listed under this 2 topics:
- search field doesn’t work well – seems that the search for widgets by typing the desired functionality doesn’t work. Since it doesn’t work, people try to guess the widgets names and type it.
- The categories options are hidden
- The category of a widget isn’t always the right one – the “most used” category, though, is used and seems to please everyone
- There isn’t much information about a listed widget

Within this topic, when discussing about the “categories” issue, the concept of tagging widgets came up. [notmart] was the first to bring it up when saying

categories are not much descriptive because there are just one, a mechanism like tags would be more neat.

A couple of people agreed strongly with the ideia of tags, but the idea wasn’t develop within this topic anymore.

When discussing the last issue of the list (not much information about a widget), [cmarcelo] brought up the following:

I think having more information about the plasmoid (or at least make it possible). Having an easy way to preview (screenshot or live preview) directly on the list… This makes discovering easy IMHO.

This statement brought a whole other discussion.
It can be listed a few topics discussed about this issue:
- Live preview of the plasmoid can be quite challenging
- Preview could be a requirement for widgets developers. “provide a screnie of your plasmoid” [pinheiro]
- Screenshots would also solve the problems and doesn’t seem really hard to get it
- We could have both: screenshots and live previews.
. If the developer provides a screenshot, use it, if he doesn’t, use live preview [jeez__]
. Use the screenshots and maybe a desktop file config key that says it supports live preview [notmart]
. Screenshots of the already used widgets, so it would be more recognizable to the user

[seele] brought, then, the important question:

where would that screenshot be displayed?

Some discussed using the previews instead of the icons, but it was noticed by [seele] herself that the thumbnails area is too small and the preview would lose a lot of details. It could be, though, exposed on a mouseover tooltip – a little larger tooltip.

This topic was finished here, but, at the end of the focusgroup session, [aseigo] entered and made an important statement about the preview/screenshots matter:

however … screenshots can be hard to see at small sizes and decent icons tend to be just as effective”, “and esp as themes change, i’d hate to have to update screenshots all the time.

8 – What usability problems do you feel about the current widget explorer?

The first topic brought in this topic was the idea of hiding the widget explorer on drag and drog, so the plasmoid could reach more space of the desktop on the drop. The idea seemed to please most of them.
The second issue brought was the confusion about the contextual controls and global controls of the “add” and “remove” widgets. [seele] made a good statement about it:

Currently you select an item from a list and click the add widgets global button, but to remove it, you use the contextual remove button in the item. The separation of complementary controls is confusing – you would expect the controls to be grouped together.

That means: to add, you click on the add widget in the bottom, and to remove, you click on the “-” sign next to the item.

Still about the add and remove topic, it was noticed that the “-” sign is used as an indicator of how many of that widget is being used, and also as a control to remove the widget (which, by the way, removes all widgets at once but doesn’t made it previously clear to the user [MoRpHeUz]).

9 – And what about functionalities issues?

The installing and uninstalling new plasmoids turn out to be a major concern – participants claimed it is painful to do.
Suggestions were about more integration with gnhs and a common interface, not evolving more then 2 steps to install and use -

i.e. launching their applet browser from the applet browser and then the applet you picked doesn’t go directly on desktop but appears on the main appletbrowser list… [notmart]

Still within this topic, ideas of rating and commenting widgets were brought. Comments/rating appearing in something like the tooltip, as the preview tooltip idea. This ideas probably came to cover the necessity of more information about the applets. Another idea, to sum up, was: a single click on the widget would expand the item and show the new informations about it: preview/screenshot, comments/rating.

14 – Take a look at these mockups and talk about it:
http://www.kde-look.org/content/show.php/Plasma+-+Add+Widgets+Mockup?content=101423
http://blusrcu.ba/nookie/?p=22

This topic can be described in topics:
- Long&linear list, mac os like, doesn’t solve the problems
- Something that can show more plasmoids is needed, cause there are lots of it
- Mockups show widgets explorer attached to the panel – seems not nice
- Window become opaque when drag is started

Untill now, the item “window become opaque” was the only one with approval of the most of the participants. The other issues weren’t solved, but brought us something to think about, indeed.

Some ideas about the way widgets are displayed came up, but that wasn’t a solved discussion too. The ideas are these to be considered:
- Multi-row thing. Concern: even less info appearing [notmart]. Possible solution: auto grow on hover [pinheiro]
- Tooltips
- Meta info pane (dolphin like) with screenshot, etc. [pinda]
- Categories in a side panel always on, instead of combo, or something like a tag cloud [pinda, notmart]
- Horizontal scroll list, taking advantage of the screen width [aseigo]

————-

At the end of the session, [aseigo] entered the channel and did a strong sum up of his point of view:

well, it’s probably all been discussed, but what i really would like to see is: * quick navigation, *categories, * search, * beauty, *simplicity”, “> install/remove is the other major feature set needed.

He also brought the question about the need of having the remove option:

removing applets that are running == not very useful anymore.

If there were more people at the time, this would surely bring lots of discussion.

————-

So this is it! As you can see, the focus group session gave us lots and lots of material do work on. It was 2 hours of good talking

Thanks again to everyone who participated and helped this matter.

annie.C

GSoC – Phase 1 – A Plasma Focusgroup

2009-04-23T02:53:49+00:00

Like I promissed you, the step-by-step of the GSoC project will be all described in this blog.

So, for now…

As I said in my proposal, the first task of this project, besides studying libs and stuff, would be the appliance of the focus group technique, right?! Ok.

But, first, what is this focus group about? Well, it’s something like a meeting with 8 – 12 people with opinions (very important – they gotta have opinions) about certain matter. There is also moderator that is going to lead this 1-2 hour discussion, and he will make sure that all the points and questions to the matter were covered. Here is wikipedia to help you: http://en.wikipedia.org/wiki/Focus_group

In this project, the moderator would be me and the matter would be the Widgets Explorer, of course.

But how to get it done? Well, the traditional technique says it must be a traditional meeting, with people in the same room. But since this matter involves a international community – and so it would be hard to get together at the same room – how about turning this room into an irc channel at freenode? That would be much better, huh?

Ok, so here is the real deal:

Me and Rob Scheepmaker (a.k.a. pinda), another student at GSoC with the Network Transparecy in Plasma project , are going to organize a focus group at #plasma-focusgroup, on friday, at 1 PM, UTC timing. Me and pinda are together on this since we are both dealing with applets.

*Just to get your attention

There are going to be 12 people (hopefully) discussing the applets subject and me and pinda are going to be the moderators of this focus group session.

Don’t worry, we’ll be following a script so that the discussion won’t get lost. We are also going to try to make things not messy, with the help of the participants

Please, if you are willing to help and want your opinion on the discussion about the applets and widgets explorer, send me an e-mail (anaceciliamb@gmail.com) with your name, relationship with the community (designer | developer | user), and registered nickname at freenode – then I can put you on the invitation-list of the #plasma-focusgroup channel.

We’ll try to gather 3-4 designers, 3-4 developers and 3-4 purely users.

The list has already begun:

ivan_cukic (my mentor!)
aseigo
nookie

Please, if you want to help, email me!

Thanks you all!

GSoC, here I go!

2009-04-20T22:50:07+00:00

Hey everyone,

This is my turn to start posting in english. So here I go!

This post is for the great news – the next post I’ll apologize for the careless with my blog and explain the reasons.

And for the great news: I got a proposal accepted for the Google Summer of Code 09! Great, huh? Man, that feels great indeed! From now on you’ll be keeping up with the step-by-step of the project from here, starting from now!

My proposal was for a new Widget Explorer (add applets dialog) to the Plasma project with the KDE institution. For a better understanding, here is my proposal :

Motivation for Proposal / Goal:

Plasma’s soul resides on Plasmoids and so it ought to be treated with love. An important concern about the delivery of these applications to the users should be “how do we do it – how do we deliver such applications to our regular users, so that they can make the best use of it?”, “how to they like it?”. These questions have been made and the answers are not all satisfactory: plasma’s widget explorer, used by most of users to search for their best suited applications, is useful, but not suitable.There’s a whole world of widgets going on in there, and also another whole world on the internet – this much of information and treasure should be pretty well organized and delivered carefully planned, aiming the best experience and use of it.

That’s what’s the Widget Explorer does: let’s us choose among a world of widgets the ones that suits us best and make it possible to easily add to our workspace.

Today’s widget explorer, though, gives too little information about what the applications are about and how it should behave in your machine. Also, there is no concern about the worst case scenario, where a user has dozens of Plasmoids and no way of simply organizing them. Today’s designers have too much work to do, since they’re the ones who do all the plasmoids icons to the explorer. What if we could provide something to represent a Plasmoid while it has no icon yet? Or maybe no icons at all? Some say anything is better than what we have today – so I’m willing to make this something come out nice!

The main motivation of developing such project is the fact that it’s clear to me the importance of this interface to the plasmoids world – it connects the user to a whole world of new possibilities and opportunities. Being aware of this matter to the KDE/Plasma community, I decided to propose a new interface to be used by hundreds of users every day, and deliver it beautifully – modern, plasma-like, with better organized information, intuitive and beautiful.

Implementation Details:

The conception of the project begins with a user centered study where I’ll try to find the exact problem with the new interface and then try to come up with the best suited idea for the New Widget Explorer.

The user study, the idea conception and the mockups will begin with the focus groups, which is a technique where I’ll group a bunch of people (users/not users, developers/not developers) and analyze their experience with the widget explorer. With that, I will learn what are the most common errors and dislikes. After this activity, I will do an analysis of similars (to see what’s going on out there) and right after that it’s time to start thinking about the application – that will be done with brainstorms over the product briefing. As a product of the storms in my brain, I will deliver mockups to the community, looking for its approval.

I won’t forget, though, that there are a few mockups made by the KDE community that are available as suggestions to the ‘Add Widget’ dialog – I collected some links that ought to be considered on the interface prototyping phase, since there was already a study made over this matter that must not be discarded:

Add Widgets Dialog Mockup

(I just want to emphasize that the whole user study, idea conception and mockups are going to be made at the weeks before the official start-coding week. It’s an extra-necessary-work that I want do to in parallel with the libs and codes studies.)

And now it’s time to code!

Since the community is not happy about the today’s explorer and there would be too much to be done to make it nicer, I decided to re-build it from the beginning and make it even better. The backend is going to be remade, with few reuses, and its structure will be made according to the new functionalities.

Today’s ‘Add Widgets’ dialog will be discarded in favor of a brand new one using the power of libplasma. The new dialog will be carefully designed, keeping in mind the usability and the Plasma way of doing things’.

Apart from the new and more specific functionalities that may come after the user study, there are functional requirements that can be already listed and which will be implemented:

- Ability to easily select and add a new widget;
- Ability to know what the widget is about;
- Ability to do a fast and easy widget browsing, even on a large collection;
- Browse over widgets shown in categories;
- Ability to install / remove widgets:
.Downloaded plasmoids;
.From local file;
- Downloaded Google gadgets;
- Ability to know whether a widget is active or not.

As you can see, my work starts this week already, together with my lib studies.

This week I’ll be doing the focus group technique, which will be applied offline, with local people – it will not be an interview or questionary, keep that in mind. The focus group will serve as a guideline to the process of the interface conception – as well as the mockups I gathered at the community.

My timeline looks like this:

Week -5 to 0 (April 20 to May 23)

- Study of libplasma and familiarization with the widget explorer code

- Focus Group

- Analysis of similars

- Product briefing and the list of requisites tightly raised.

- Brainstorms, brainstorms, brainstorms

- Delivery of the mockups to the community for approval (especially designers’ approval)

Week 0 (May 23)

- Start coding and constructing the skeleton of the backend of the project:

Week 1 to Week 4

- Build the whole backend necessary to:

- Deliver the Plasmoids to the users and make it work on the desktop/panel;

- Deliver the functional requirements of the Widget Explorer;

- Deploy the specific requirements that will be collected after the focus group activity. That means: if tagging/grouping the users’ plasmoids is necessary, I’ll develop the backend with an optimized way of recording such informations and listing the plasmoids according to these informations;

- Discard the need of Plasmoids icons. That means: if instead of icons there will be a screenshot of the applet, or even an actual preview of the thing, the backend needs to support it.

Week 5 (July 6)

- Create a simple supporting GUI to see the backend in action and run all the necessary tests. Make it all bug-free.

Week 6 to Week9

- Start drawing the necessary svg’s to implement the GUI

- Start coding the final GUI with libplasma and, if necessary, implement extra animations.

Week10 to Week11

- Integration between the backend and frontend.

- Run the final tests and fix the final bugs

Week 13 (September 3)

Submission of required code samples to Google

This is a big thing for me, and hopefully you’ll see a great project being made!

Now let me go celebrate!

cheers!

Canola Layout: so Qt!

2009-04-03T14:35:39+00:00

Since the first time I saw Patifa’s presentation, I was very impressed about that Canola’s menu layout discussion. I’ve been a user of Canola since its first release, but had never thought about how hard it would be to implement such a dynamic layout.For those who don’t know or remember what I’m talking about:

The first column on the picture shows what the designers wanted from the layout’s behavior.

Here at OpenBossa@INdT, we are working on the Animated Layouts (there is a previous post, in portuguese) with the Qt Kinetic project, as you should know. On this last sprint, we decided to code some demos showing new usages of our QGraphicsLayoutProxy (QGLP), different from the previous ones, to test the new API we came up with.
And we did it.

A custom layout example, Qt Canola Layout, which demonstrates how to use QGLP “implicit” on an animated layout. Just add your widgets to this custom layout and let it handle with them! Check out the video in Openbossa channel.

We spent some days talking and brainstorming this algorithm. If you manage to check the engine’s implementation, you will see that it was a quite good work. The secret here was to find out what proportion designers were thinking of, and how to implement a ratio for that. The golden ratio was a good starting point… ;)

Maybe I’ll post about this engine again, goind deep on it. For time, don’t forget to check the latest Bossa Conference 2009 videos on our other channel, on blip.tv.

Also, check Anselmo’s post about this example and on a previous one, about the integration of Qt State Machine and Animated Layouts.

jeraman

2009-04-01T20:19:33+00:00

pymt is probably the easiest and powerful single tool you can use to build multi-touch (mt) applications! according its site:

PyMT is a python module for developing multi-touch enabled media rich applications. Currently the aim is to allow for quick and easy interaction design and rapid prototype development.

i started to experiment with it while was wondering about which technology to use in my proposal for GSoC and i just fell in love! these are basically the reasons:

more than a library, a framework: big mt application are complex. they impose you to handle with lots of issues (tuio stuff, graphics, gestures, enabled multi-touch ui widgets and others), what generally would demand a different library/tool/whatever for handling each one. however, pymt already takes care of it for you! so you can just use all pymt these funcionalities without worrying about differents sintaxes or learning differents libraries.
no necessity of simulators: to test mt programs in a fast way, people generally use simulators. they allow you to test applications without necessity of testing it in a mt table. there are some good simulators available (reactivision’s one and the ashish’s one, called qmtsim) but both have to be downloaded separately and sometimes are just too complex for simple tests (moreover, if you wanna use them you’ll have to work a new window, different from the application one). in pymt, you can simulate fingers by clicking mouse buttons, even in simple applications.
easy to program: well… pymt is python, you know? (i think that’s enough… if you didn’t get the point, you should start learning python now).
fast execution – well… pymt is python, you know? (2)
multi-platform – well… pymt is python, you know? (3)

despites all these qualities, the project is just starting (the second version was just released). however, it’s stable enough to starting coding with it!

try it yourself! check here how easy it is!

further informations
[1] about pymt
[2] pymt on googlecode
[3] about python

Tagged: multitouch, python

jeraman

2009-03-25T15:41:01+00:00

are you a student? do you enjoy programming? open source? and what about money? well, if you answerd yes to all of these questions you’ll really love Google Summer of Code (aka GSoC).

“is a global program that offers student developers stipends to write code for various open source software projects. (…) Through Google Summer of Code, accepted student applicants are paired with a mentor or mentors from the participating projects, thus gaining exposure to real-world software development scenarios and the opportunity for employment in areas related to their academic pursuits. In turn, the participating projects are able to more easily identify and bring in new developers. Best of all, more source code is created and released for the use and benefit of all.”

we at wouwlabs are working hard to get this done. in my case, i’m trying to work under Natural User Interfaces (aka NUI Group) mentorship.

NUI Group is an interactive media community researching and creating open source machine sensing techniques to benefit artistic, commercial and educational applications, and the greatest group in world about multi-touch technology. It was also one of the biggest collaborators to our success in building our multi-touch table (here, link in portuguese, or here). it would be a real pleasure to reciprocate them and colaborate with NUI’s growing!

you can read my idea proposal here. any feedback is welcome!

further informations

[1] about GSoC 2009
[2] about NUI Group
[3] my idea proposal

Tagged: eventos

imagem41

2009-03-22T01:09:16+00:00

this post will explain how to extend Pd functionalities to other languages, allowing us to use it something like a API. with this, we could for example create a [osc~] and connect it to a [dac~] using for this just a python program. this is possible due to dynamic creation of Pd objects (aka dynamic patching) realized through sockets.

it’s important to notice that this is different from simple exchanging values of variables through sockets. if you want this, i advice you to take a look on netsend and netreceive objects.

to make things more clear, i’m going to use some examples. they can be found through svn at:

http://jeraman-lab.googlecode.com/svn/trunk/dynamicPatching/

1. putting pd to work without gui

well… that’s it! this simple step makes Pd run without any gui. you can do this running pd in command line with -nogui parameter. to illustrate how we can use it, execute remote.pd file according the above. after it, just execute normally client.pd and see what happens.

running pd on -nogui mode

this is important if you dont want to see any pd patch appearing on the screen during the execution of your non-pd program. if this doesn’t make sense to you, just skip this step.

2. dynamic creation of patchs and objects

this second step is about dynamic patching itself. it consists on sending messages to pd, that interpretes and tranfosms them into pd objects as soon they are sent. these messages are sent by [send], what means that we could send them through socket from another program (we’ll do this in next step). you can notice that the messages are almost the same to the lines you see when you open a pd patch with a text editor (if you never did this, try it!).

sending the messages to pd

to see how it works, just open sample1.pd. the second file, sample2.pd, is a little bit more complex, with more objects. if you need to create others, i advice you to read damien’s tutorial or yet put the desirable object on a pd file, save and open it in a text editor (remember: the messages are almost the same).

3. the mix!

now, it’s time to make the two steps above together! the stuff works almost like i already described, but now instead using [send] object, we’ll use pdsend file (available in pure data directory) to connect to pd, or yet a simple tcp client. despites, we will initialize a simple server (it contains just a receive object) in -nogui mode. this is the patch we are going to modify dynamicly, so it’s our entrance to pd program.

the python program that creates pd objects dynamically

to show that the stuff really works, i wrote a simple program in python that allows you to create and connect pd objects. to put it to work, just open it and change the given place to where are pdsend and server.pd (variables pdsendDir and serverDir) files.

that’s all! i hope this could be usefull!

thanks to: jarbas jacome, for his tips, and damien henry, for his brilliant tutorial about dynamic pathing!

further informations:
[1] about puredata
[2] damein henry’s tutorial about dynamic patching;

Tagged: pd, tutorial

jeraman

2009-03-16T00:44:38+00:00

last post i talked about quartz composer, a apple technology focused on graphics that i’m using in a project. in spite of the compliment i did, in this post i bring bad news…

first of all, i wanted to use osc protocol with qc. after some search, i found a excellent post about the theme at rubberhawk’s blog, that in fact really works. it made me think that things were going to be easy… but they were not.

as soon as i tested the osc patch i noticed that something was wrong. in my project, there’s a tracking module, made with openframeworks, that sends through osc the coordinates of the tracked object. there is no problem with it, but the qc module that received this information sometimes had its signal “broken”, and all my variables turned to zero for some miliseconds.

i got very confused with this incomprehensible problem, but – after some hours of useless search – i found out: yes, it’s a quartz composer bug (you can check it out here): no one knows what exactly happens, but the fact is that osc quartz objects have some problems… i really hope that apple could fix it soon!

in my case, i began to look for multicast udp communication and this quite fits my needs. who knows it doesn’t fit yours? (check this link).

ps: by the way, the guy that wrote this is a genius.

Tagged: osc, quartz

jeraman

2009-03-11T22:11:50+00:00

quartz composer (qc) is a visual programming enviroment largely used by VJs (mainly with others softwares like VDMX5) freely provided by apple, making part of osx development enviroment (unhappily, just for this plataform…) and focused on rendering and processing graphical data. it looks like vvvv for windows, but i think that it’s more clean and organized (a more detailed comparison, here).

picture of a qc code and its equivalent program

the first time i heard about it was some days ago, when braga and i were studying mapping techniques. after reading some nice tutorials (check these out: here, at mac apper, here, a youtube video by irishman, and here, a vimeo video by Christopher Wright – all strongly recommended for those who want to learn qc), i felt that i was able to start coding. i did it and, not a surprise, it was very very intuitive and easy. something arround an hour later, i had my first program: a crazy cube whose faces i mapped images from a webcam and spins according the sound that a microphone gets. here’s some shots:
so, if you are on a mac and need good graphics, quartz composer is certainly a good try.

ps: in fact, we are now using qc in a project called kaly yuga. i hope to talk more about it soon =D .

further informations
[1] about quartz composer
[2] site with lots of intersting samples for quartz
[3] about vvvv
[4] differences between quartz and vvvv

Tagged: quartz, study

projection mapping + vvvv!

2009-03-05T21:29:11+00:00

Há um tempo, quando vi uns vídeos do AntiVJ (técnica, live painting e building projection) sobre Projection Mapping, fiquei muito instigado com a quantidade de possibilidades que aquilo abria e com o impacto visual. Naquela hora aquilo parecia muito distante, pois supostamente utilizava técnicas obscuras e muito avançadas de projeção, de forma que eu mantive minha cabeça com outros focos, vez por outra pensando em como seria bom dar umas projetadinhas por aí.

Ontem, eu e jeraman projetamos. E, como acontece com a maioria das coisas que experimentamos, não é nada de mais. O vídeo abaixo tenta mostrar o que fizemos ontem: a gente arrasta os pontos e especifica onde queremos projetar. O efeito que aplicamos foi o de uma luz que rotaciona ao redor da caixa, mas as coisas podem ser BEM mais elaboradas (vide links de vídeos acima).

A técnica é bem simples. Para direcionar a projeção, você projeta o que quiser onde quiser e “projeta preto” – ou seja, não projeta - onde não quiser que nada seja projetado. Utilizamos o vvvv, um framework bastante interessante para trabalhar com síntese de vídeo/animação/áudio em tempo real. O legal dela é que é BEM fácil de usar e MUITO modular. Mesmo para quem não é programador. A programação é totalmente visual, daí que nós não escrevemos If-elses e whiles, mas simplesmente ligamos caixinhas. Além disso, muita coisa já está pronta para facilitar nosso tratamento de câmera, iluminação, interação etc. E mais! Você quer uma partezinha de outro projeto que lidava com, digamos, a fonte de luz? É só dar um copia-e-cola que você importa para o seu projeto na hora, depois é só ligar os fiozinhos. Se você não entende uma “caixinha”, dá um F1 que ele abre uma janela explicando e ainda com uma aplicação-exemplo.

parte do codigo VVVV

Daí, mais uma para o saco das técnicas-super-empolgantes-aparentemente-complicadas desmistificadas.

O código tá aqui, mas no próprio site do vvvv você pode achar muita coisa que o pessoal coloca.

jeraman

2009-03-05T16:20:36+00:00

it’s been a long time since i talked in my old blog about antvj and mapping technique (link in portuguese). in that time, i had no idea of how they could do that awesome stuff. in that time.

using vvvv (like pd, it’s a visual programming language, but focused on graphics), braga and i started to play with the technique. unhappily, we still don’t have such incredible thing that they do, but we are studing!

further informations and explanations at braga’s blog (link in portuguese). that’s not dificulty, in fact it’s very easy (more than we think). check it out!

next steps: automatizated detection of edges and lines, in antivj style! maybe bisoiando (a python module about shapes tracking that i wrote sometime ago) could help us!

further informations

[1] about mapping technique (here – on youtube – and here – on the vvvv wiki)
[2] about antivj
[3] about vvvv

uCon and wouwlabs’ workshop

2009-03-01T19:38:41+00:00

Yesterday I was at uCon Conference. In just some few words: it rocked a lot. I can’t remember last time I went to a security conference so well organized! (at least on Brazil…) The lectures, the speakers, the people, the discussions, the food and the workshops were ALL awesome! I’m really proud of being part of that. :) Not even ITA’s SSI, when it still existed, has ever managed to do such a great job! Thanks to H2HC guys and their 4 years of experience!

The conference happened on a restaurant called Jardins, at Recife.

Besides the presentations, there were 2 workshops. One was a great workshop titled “Introduction to Electronics” (leds, ir, pov, resistors, pic’s, etc) and the other one was a wouwlabs’ workshop called “Tinkering, DIY and Multitouch 101″. It was amazing and people seemed to enjoy it a lot! Actually, I wasn’t directly involved within it this time, so congratulations to annie, braga, calegario and jeraman! You all did really great! ;)

Lost it?! Don’t worry, there will be others. ;)

Life is random.

braga, anniec, me and calega

2009-02-27T22:18:52+00:00

we are happy to announce that wouwlabs will lead a workshop at ucon security conference about diy, tinkering and multitouch technology!

uCon is a vendor-neutral and single track conference on hacking, technology and information and telecommunication security. it’ll be held in Recife, Brazil, and starts tomorrow!

ps: the workshop (as well the conference) was a great success! here goes some photos:

ucon security conference

the tinkering mess

anniec

uhu!

calega and braga

braga

the audience

braga, anniec, me and calega

Tagged: others