This week I did my exam for ‘Diploma de Estudios Avanzados’ – short DEA. This marks roughly 2 years time spent in Spain doing my PhD. The “exam” is fairly simple, basically just a presentation about the research I did during the time and the courses I took. My presentation was mainly about the Semantic Web research I’ve done and the areas I apply it. You can check you the slides here.
I was talking about:
- research I did on mashups during my first year here and the courses I took concurrently on the university (all this was rather a brief talk)
- my current research on applying Semantic Web to Idea Management and my project (GI2MO) that will hopefully be the main pillar of my PhD (this was the main part of the presentation)
After being awarded with DEA the road to getting the PhD diploma is opened and in theory, if I had done enough research/publications, I might as well get it the day after (which obviously won’t happen ;p).
For quite some time now, I stopped using Firefox
and switched almost completely to Chrome
. Unfortunately, the letter has much less plugins, including no solid plugin to detect RDFs linked inside html documents
In Firefox Semantic Radar used to do the job but I got tired of running both Firefox and Chrome when i was working with RDF so….I’ve written my own plugin for Chrome – RDF Detective 🙂
It works almost identical to Semantic Radar, in documentes loaded by the browser it detects:
and displays a little clickable icon to open RDF in an external viewer (you can edit the viewer in the plugin options by default I’ve set it to SIOC Browser).
Hopefully, when i get more time I’ll hack the plugin to have more preferences and add few other things. For now, it does its job good enough to use (for me), this is how it looks:
At the moment, I’ve published it only on Google Extensions website, if you feel like testing, head over there and install it. So far I only tested how it works on Mac Os X, running Chrome 5.0.342.9 beta, so feedback welcome!
Update: tested for Windows7 + Chrome 22.214.171.1244. Works as well!
Update2: New version out! RDF Detective 1.3 now supports RDFa detection (renders a different icon and allows to open RDFa in a different viewer then RDF). I also added experimental support for detecting what RDFa namespaces are used and in relation to that render the icon in different ways. Currently 3 options: if there is only Facebook or Opengraph namespace their specific icons get rendered, if there is more annotations of other namespaces then a generic RDFa icon is shown.
Update3: The source code for the extension was published at: http://github.com/cyan-ide/rdf_detective
FOAF is a data schema for describing personal data on the Web and is one of the most known in the Semantic Web world. Basically, its a way to publish information about yourself on the web in form of pure data, just like im doing on this webpage of mine but readable for machines! How to do it and why ? Read on!
Part A: The Tools
So far I’ve only had a FOAF file hosted at my old workplace but now I thought that like every normal(tm) person I should put it up on my homepage. This time tho, apart of the regular personal details I wanted to include all the other data from my page so: publications, projects, link to cv and so on. Simple ? Not as much as you might wish. Here are some tools I’ve tried to make it faster then writing the RDF code by hand:
1st try: Protege. Import the foaf ontology, make the instances and fill out the forms, and thats it! …..but wait not really. Its not too user friendly even for somebody who knows more or less all the Semantic Web technologies…
2nd try: TopQuadrant Composer. Same story, import the ontology, make the instances and…damn its exactly the same story…just a little bit less confusing.
3rd try: Dedicated user friendly apps. So I abandoned the ontology design tools for more simple stuff. Ye some work(tm) and some are even quite nice looking but they kind of don’t let fill out all that I wanted (read projects, publications..). So maybe some other time.
4rd try: “reuse, reuse, reuse”. So finally I’ve decided not to trouble myself with any tools and go back to the stone age and write everything by hand…well almost.
Recipe for writing your own FOAF profile fast(tm): google out a known (Semantic Web) person/geek , download his FOAF file and modify it in an xml editor. It took 30min vs. 3 hours or so trying to find a nice app 🙂
Here’s my profile.
ps. I used Wikier’s (aka Sergio Fernández) profile to make my own. Cheers mate!
ps2. If you are waiting for a earlier promised “why to make your foaf profile”. I have to disapoint you. If you are a regular person that does not have much to do with Semantic Web, then to my knowledge there is absolutly no reasonable practical cause why one should make his FOAF profile…not yet at least.
Soo finally proposal time ended and I’ve started doing some initial research towards my PhD that I plan to do on idea management techniques (and few other things that I wont disclose now :p).
The first goal for me is to find all out there and analyze the state of the art in both research and industry. So far the academia part seems fairly easy since there is quite little public research on the specific topic of idea management (like software rather then methodologies). The industry study is the real pain!
It’s been over a week of reading information on the web and to be honest at the moment I think it might be actually easier to forget about a deep SoA investigation and construct a innovation environment from scratch with my ideas rather then investigate all solutions out there available on the market. When at the end of the day I’m happy to comprehend all, I get to find the next day another 10 apps that claim to do the same and be ‘the market leaders’ 🙁 Also the amount of marketing flavored info stead of facts makes my head hurt.
Oh also dont bother sending requests for trials or demos to companies in the area…responsivness is 0.
Either way I’m starting to write my report on everything and publishing it hopefully fairly soon!
Finally the EU proposals rush is over. It was a great time with lots of new experiences but now finally I have some time to catch up with other things like doing actual research or .. writing on my scientific blog:)
During October I went to WISE 2009 conference hosted in Poland/Poznan this year. I was presenting a paper on mashups and integration of visual environments:
“Integrated Environment for Visual Data-Level Mashup Development”
here are the slides of my presentation: mashups_wise2009.pdf
The presentation went surprisingly well, and instead of doing a 20 min talk, I think I did 35-40 min since one of the presenters after me didnt show 😀 My paper was really short/simple and more like an idea rather then genuine research (initially I got accepted for a workshop which was then converted to a special session or something). I kept the presentation simple as well and to fill the rest of the time I tried more of a business oriented talk so I had lots and lots of questions and feedback from the audience. Feels nice ! Ok maybe it was less science in all this but I think it was more enjoyable for everybody and I had really a lot more options for collaborations afterwards then when giving the usual geek talk. So there was time for details in f2f talks too. Too bad this is not really my primary research area. In general – good training for presentation skills but I think I’m happy that the paper didnt get accepted for the bigger conferences before, they would eat me alive I’m sure of that 😉
As for the conference – I’m a bit disappointed with it, did not really find much interesting presentations for me during the conference time and well…after WWW earlier this year, I think its gonna be hard to impress me anyway 🙂 For WWW I would be happy going just to listen, for WISE – not really, especially not for this price.
Some of the papers that had quite good presentations were during the Visualization and User Interfaces sessions by a group from Kyoto University – there were few papers by those guys mostly everything connected to mining and presenting or searching information/news about the past. Quite interesting I have to say especially that I talked with one of postdocs from Kyoto about this ealier and I knew what’s the deal.
Oh one more thing that I obviously was most interested in: semantic session – not so semantic at all and felt like listening to people from past era of Semantic Web. I suppose working at DERI and participating in events like LOD@WWW sets high standards 🙂
So while building this page I figured I wanted to have some kind of easy tool to maintain my publications and projects lists. After some searching I found 3 interesting WP plugins:
- WP-Table Reloaded – adds a nice looking GUI in admin panel to create and manage tables
- bib2html – generate html directly from bibtex file and put into WP page (unfortunately discontinued and not working in WP2.8)
- BiblioFly – generate bibliography from its internal base (each resource has to be entered by hand with a GUI in admin panel)
Out of the thee, WP-Table is apparently the best maintained plugin but its not directly aimed for generating the bibliography. Nice GUI, lots of options and CSS customizable.
On the other hand, Bib2html appears to be the best bibliography dedicated solution (especially when used together with i.e. Zotero to organize bibliography). It has good features but hardly any GUI and unfortunately it doesn’t work in the latest WP (2.8).
Than BiblioFly is something in between: like WP-Table has a GUI and internal database but its only aimed for bibliographic references (without bibtex import option like bib2html) and seems in a bit worse shape then WP-Table.
More to come when i decide which one to use.
So…one year passed since I came to Spain. As expected, it was a major change, especially that I’ve decided to finally start a PhD after working for past years in Ireland at the DERI institute.
My first year on the university was in large part about participating in obligatory PhD courses (in spanish!). Now, I’m almost done with that (the student year formally ends in September). For each of the courses students have to prepare some kind of work – usually a publication style paper and present it in class room for others. Depending on the course type I was writing and investigating some interesting topics. I’ve put quite some work into this, hence, I figured I’ll publish the papers here:
- Semantic Web evolution: towards the usability of data (for Aplicaciones y Servicios Avanzados Internet course) [paper]
- The Two Cultures (for Aplicaciones y Servicios Avanzados Internet course) [presentation]
- Overview of Cloud Computing large-scale processing technologies (Metodologias y Arquitecturas Orientadas a Objetos para Sistemas Telematicos) [paper] [presentation]
- Dynamic FAQ systems (Agentes Inteligentes. Aspectos Metodologicos, Logicos y de Contexto) [paper] [presentation]
- Sentiment Analysis: Introduction and State of the Art overview (Tecnologias Lingüisticas y Aplicaciones en la Web) [paper] [presentation]
- Metodología de desarrollo ágil para sistemas móviles (Temas Avanzados de Ingenieria de Servicios Telematicos) [paper]
- some paper about PhD Thesis planning (Metodologia y Documentacion Cientifica) [paper eng] [paper esp]
- bunch of writeups about seminars on UPM during the year (for Seminario de Investigación course)
Yep, it’s a fact. After quite some years of my engagement in various research projects, I’ve finally decided to run a research blog and publish information about various things I do or think about with relation to my scientific work.
Time will tell how sucessful I’ll be with this but I plan to update this space constantly and publish the progress/news of my work on PhD. Also most likely I’ll put here some ideas regarding my work, since lately I’m getting a lot of those and I figured it’s part of my work to share this too!
Welcome and enojoy.