User:Guslacerda

I am a person. I have a website.

Disclaimer: Strictly speaking, I am not a native speaker of English. For most purposes, however, it is my best language. It would make little sense for me to list Portuguese as 5 if my English were listed as 4. I would like a more objective way of rating one's language skills, such as Cloze tests. Also, my Dutch should be a 3.5. See User:Guslacerda/Languages

My contributions to Wikipedia
contributions as Guslacerda When not logged in, I sometimes edited as User:84.81.60.126 (contributions) This page's history suggests that I've also edited as: 128.237.225.179, 128.2.188.52, 151.201.247.135, 151.201.233.133, 67.171.70.182, 76.80.9.155, 128.2.247.105, 128.2.247.28, 201.52.64.39, 201.9.134.208, 146.50.8.81.

of academic relevance (very broadly speaking)
Alan Bundy, User modeling, Marc Hauser, Joseph Halpern, Production rule, Model (abstract), Folk science, Independence-friendly logic (here), Bayesian cognitive science, Many Eyes

Brazil
Frevo, Mombojó, Frans Post

Angola
Capanda Dam

Music
Jason Carter (fiddler), Mike Auldridge, Alison Brown,

Other
Panarchism

Less significant contributions, but possibly interesting:
I added the 5-string banjo to Template:Vocal and instrumental pitch ranges, Validating carrier, Matt Ridley, Exposure effect

Admirable Wikipedians who I've met in real life
User:Kzollman, User:Michael Hardy, User:Metaeducation

My thoughts on Wikis
I believe that Wikis need:
 * editing in-line, both for real content and comments (in normal wikis, once you hit 'edit', you lose your visual bookmarks of what needs changing). See also the WYSIWYG issue below.
 * modular design, à la software design. MediaWiki offers Transclusion in the form of so-called "templates"). I am interested in general design principles, whether applied to scientific methodology (see Lakatos) or software development methodology. See, for example, 1 and 2).
 * more formal structures: interactive reading through semantically-rich text. I like the ideas of the Semantic MediaWiki.
 * Integration with endorsement-based reputation systems, so that you know how much the information can be trusted. See TrustOPedia, my system project for a practical epistemology. In the same vein, Wikipedia's history viewer should allow one to query which editor (i.e. which version) introduced a given piece of text into the article.
 * Smart-linking: renders images that give information about the linked page (e.g. how complete / how rough the link is / whether it gives academic references, etc) without committing to loading the whole page.
 * integration with LaTeX for rich editing (see WikiTeX), and hybrid editors that mix WYSIWYG with text (à la DreamWeaver).

Some of my thoughts on this are influenced by User:Metaeducation.

My website is based on MediaWiki, except for the home page. I haven't implemented any of the above hacks.

My Wiki Sandbox
WikiTeX seems to work here
 * $$ \emph{M} \vdash P(x,n) \rightarrow \sum_{n=0}^\infty \frac{x^n}{n!}$$

$${\bf e} = x$$

$$B_{M} = \left[ \begin{array}{ccccc} 0 & \ldots & 0 & b_{k,1}\\ b_{1,2} & 0 & \ldots & 0\\ 0 & b_{2,3} & \ddots & 0\\ 0 & 0 & \ddots & 0\\ \end{array} \right]$$

Early years (1979-1997)
I was born in Recife on 11 Feb 1979. I loved the beach in Boa Viagem and had an obsession with geographical facts. In 1986, I got an MSX and learned to program in BASIC. In 1989 (age 10), my family moved to Luanda, as my father worked in the Capanda project, building a hydro plant on the Kwanza River. I learned to skateboard, to bike, and to play bete, a variant of cricket. In mid 1992, because of the upcoming elections with a predictable outcome (war), we returned to Brazil. 1993 was spent in São Paulo, and in 1994 we moved to London, where I gained fluency in English. Programming remained a lonely hobby (now with MS-DOS BASIC), and my projects included reverse-engineering Minesweeper and testing hypotheses about number theory. I studied at Southbank International School in Notting Hill, where I met a very diverse group of rich kids, while following the IB program, which included an independent project on fluid dynamics. An angsty teen, I subscribed to mystical beliefs, became interested in taoism and zen buddhism, until I grew out of it 3 or 4 years later (since then I've been a pretty good skeptic). My main hobby, which I enjoyed a lot, was playing soccer, both with my school, and with strangers at Kensington Gardens. I also learned to both enjoy and despise solitude. I often went to the Whiteleys mall and movie theater and bookstore by myself, and spent a lot of time photographing flowers and clouds. I once read about Laplace's demon in James Gleick's "Chaos" and it had a profound effect on my philosophy.

College (1997-2001)
For reasons still mysterious to myself, I decided to study in the USA. After visiting 6 colleges in October 1996, I ended up at Bucknell University with early decision. I started out majoring in Physics and Philosophy. My Philosophy major quickly turned into Psychology. Linguistics was not an option. Displeased with physics education (see my compiled rants), I became a Mathematics major. As the Psychology curriculum too proved disappointing, the distant promise of being able to take an AI class (looking back, I have been an "AI person" ever since my teens), along with a small bit of financial pragmatism, convinced me to major in Computer Science, a decision which surprisingly (to me) paid off in terms of knowledge of programming languages and algorithms. In the Summer of 2000, I interned at Bell Labs and met many geeks, among whom I found my social home, like I never had before (fortunately, this would happened again, in less temporary fashion, at CMU in 2006). Finally, in May 2001, I got a B.S. in Mathematics and Computer Science.

Boston and Traveling (2001-2003)
Unemployed, I moved to Boston. Spoiled by college, and with few social channels, I had to figure out how to meet people. I quickly became a regular at the bluegrass sessions at the Cantab Lounge and taught myself to play fiddle (with much help from that wonderful community). I also bought a guitar, and learned to jam with other people. This way, I found out about my talent for melody (I've been obsessed with solfege ever since I had piano lessons as a child), and near-impairment with rhythm. This year, I acquired a taste for Celtic music and gypsy jazz, and met several people from Berklee and MIT, where I often played soccer. I got a boring programming job in Woburn, which paid twice as much as I spent, despite the extra expense of buying a car for the commute (and parking). I occasionally went to other MIT events, such as swing dancing and Tech Squares. I also attended a monthly philosophical salon, and hung out with some friends from college.

Unhappy with the job, in August 2002, I left everything (i.e. my H1B, my life in America) to do something crazy: attend a logic conference in Europe and go traveling. I went to ESSLLI 2002 in Trento, and ended up doing a 19-city tour of Western Europe in 3.5 months. In Paris, I met Fare IRL.

At the end of 2002, I spend several months with my family in Recife, in which I discovered what Brazil is really like as an adult. I took 2 classes at UFPE: one on AI and one on information theory. The future was very uncertain.

Europe (2003-2006)
While traveling in Amsterdam, I had been lucky to meet the Dutch logicians again, as they were having a "schoolweek". So in 2003, I started an M.Sc. in Logic at the Institute for Logic, Language and Computation at the University of Amsterdam, what I imagined to be a sort of "Goedel, Escher, Bach studies". There, I learned a little bit about modal logic, formalized mathematics (and science), and type theory, Kolmogorov Complexity, game theory and formal semantics and pragmatics, and did work on the cognitive science of autism, automatic dictionary building (for IR purposes), and for my thesis, I wrote an equational  reasoner in Lisp (quite a learning experience!). I was part of a delicious music cognition reading group, a couple of workshops on the evolution of language, and helped organize AIED2005, on educational software.

2004 was a busy year: I started tutoring high-school students in math and physics. I met Laurens Joënsen (who made me the official Brazilian bluegrass fiddler at Blijburg, in sandy IJburg), and Ralph de Rijke. Started learning theatre improv with ProblemSolved and BoomChicago. I hitchhiked to (almost) Luxemburg on the way to Nancy. Other trips: Alkmaar, Bloemendaal (hiking), Utrecht, Leiden, Rijswijk, Voorthuizen, Alphen aan de Rijn, the Hague (Escher Museum), Nijmegen, Eindhoven.

Through the influence of my dancer housemate Erik, I started learning contact improvisation. I practiced this at the Hogeschool voor de Kunst as well as at Vondelpark. I got to know the wonderful feeling of improvised self-expression: acrobatics, dancing, singing and acting, all completely improvised! Oh, the joy of expressing pure emotion... the fun of forging semanticsless communication with strangers, and of creating something from nothing together, on the spot.

After graduating, I went to Munich to work for Cadence Design Systems as Lisp programmer. Despite the fun I had with new language (learning German is fun(ny) when you know Dutch), my social life sucked, and I felt it was time to return to academia and to America. Americans, I feel, are easier to get to know. In Germany, friendliness towards strangers is often interpreted as a sign of insanity, and after a couple of months, even I was starting to think like them (I think of it as a game with two Nash equilibria: when you think like a German, you behave like a German, unless you're insane! So when you see a local being friendly, they are most likely insane.). Maybe it's because Europeans live their whole lives in one place, whereas the typical educated American moves several times in their lifetime (which would favor the second, friendly, equilibrum). Trips: Potsdam, Berlin, Hamburg, Garmisch-Partenkirchen.

CMU (2006- )
My frustration with the non-customized format of classes and textbooks increasingly fueled my interest in educational software. In 2006, after flirting with Carnegie Mellon University's philosophy department, I came to work as a researcher / programmer for the HCI and Machine Learning departments at the  School of Computer Science, making cognitive modeling-based educational software. I also intended to pursue interests in:


 * developing argument-mapping software as: (1) a formalization and decision-making tool (2) model-tracing argumentative behavior for knowledge-tracing (a form of user modeling).
 * scientific methodology, including: (1) statistics and (2) model selection and causal inference from non-experimental data.
 * formalizing all forms of reasoning, as seen in mathematical AI: integrating multiple representations, e.g. diagrammatic reasoning
 * symbolic cognitive modeling with ACT-R.

As it turned out, I became more interested in Bayesian machine learning, and less interested in dreamy ideas like many of the above.

In July 2007, I attended a Summer School in Bayesian Cognitive Science in Los Angeles. In July 2008, I will be presenting at UAI in Helsinki.

Labels I accept

 * computer scientist
 * statistician
 * artificial intelligence researcher
 * cognitive scientist

Labels that are a stretch

 * mathematician

Labels that I might have accepted in the distant past

 * logician
 * philosopher
 * linguist

Hobbies
word games, ball games, improv games