20 Mar
2009
20 Mar
'09
6:16 p.m.
Karel Piwko wrote:
Hello,
Hi Karel,
I am student of master program in informatics at Brno
University of
Technology, Czech Republic. There is a
short list of experience I can provide to make your project better:
- Java is my primary language for more than 5 years,
including JPA (Hibernate), JTA and web development
(JSP, Struts, Stripes, Freemarker templates)
- very good experience and knowledge about compilers
- good knowledge of XML background such as XSLT, XSD, XPath and
XQuery
- basic knowledge about semantic web (RDF, OWL, Jena, Sesame)
You seem quite experienced in many domains. This is good.
During my bachelor program I defended (in year 2008)
thesis called
Native XML databases http://www.fit.vutbr.cz/study/DP/BP.php.en?id=7006
(in Czech, abstract in English), which basically was an comparison of
two portals of Wikipedia, the first based on Stripes, JPA (Hibernate)
and PostgreSQL, the other on eXist and Apache Coocon, both deployed an
JBoss application server, considering the speed (both of application and
development) and developers experience of both solutions.
There very same year I wrote fully compliant CSS 2.1 parser
http://cssbox.sourceforge.net/jstyleparser, part of CSSBox project for
semantic recognition of web pages. This parser uses ANTLR as parsing
layer.
Currently I am doing some minor work on Czech linux portal
http://www.abclinuxu.cz, which is based on Java servlets with Freemarker
templates.
I can provide my CV, references and more detailed description upon
request.
What I would like to do for XWiki?
My current focus in studying is towards artificial intelligence, which
can be applied to same heuristic optimization, but I'm mainly interested
in JSR 168, because the idea suspects good project design. Another very
interesting proposal from my point of view is Improved fetching for
XWiki Watch, as the proposal expect me to do both module desing and some
semantic recognition and content parsing and any work considering some
parsing and compilers will fit to my profile.
The portlet integration is a difficult project, since it involves
careful design in the middle of the community, and it touches core
aspects of XWiki, so you will have to prove that you are really up to
the task before getting accepted (this is a high risk project). And what
makes it so difficult is that we're converting from an almost monolithic
architecture to a component-based one, with each module designed as a
distinct build unit, communicating only through interfaces with the
other components, and unfortunately not all the components the portlet
mode needs to interact with are fully designed and implemented.
The XWatch project is not that risky, as it involves writing code with
less connections to the other modules. But it still is interesting and
challenging enough, IMHO. Anca, the XWatch leader, can provide more
information about it.
By the way, I'll be at XML Prague during this
weekend. If any developer
participates, I would be nice to discuss ideas face to face.
Sorry, but we didn't find out about it soon enough, so we can't be there.
Good luck!
--
Sergiu Dumitriu
http://purl.org/net/sergiu/