MLA citation style had to wait because some pending issues could finally be solved thanks to Neri, Ian and Anthony. Their help was like an aspirin for the "clean session" problem. As stated in previous blogs, I have not been able to find a way of cleaning session cited records once the user leaves the editor page (additionally, I wanted to store those records in database, so they were retrievable whenever trying to edit the article). I tried several methods without success until I knew through my last meeting that Joomla 1.5.4 included events onAfterSaveContent and onBeforeSaveContent. Yes, exactly what I needed. The lesson: to read the announcement page of every new Joomla version. With that information, I finally solved the "clean session" problem through two plugins: one implementing onAfterSaveContent event for storing cited records in database and the other one attached to onAfterRouteEvent to load into the session the records when editing an article. The number of lines of code required will surely make you laugh. One thing is missing: to automate its installation when installing the component as I did for the other 3 plugins, but that is a piece of cake.

Issue associated to getting component configuration parameters in backend was solved too. In that case the solution was a single line of code. Again, thanks to Neri, Ian and Anthony.

APA citation style is ready. As I told Neri, it needs a lot of testing because the amount of rules is large considering I only implemented those related to the format of cites and bibliography section. APA citation style also define rules for the document presentation in general, which are beyond the scope of JResearch (at least for the soonest release).

Parsers have been tested a little more and thanks to Neri's feedback, I identified the faced problem was related to file permissions. I should have considered that as I am a full-time GNU/Linux user. Installation routine was modified so bibutils executables has the right permissions now. However, it needs to be tested in a Windows machine because, honestly, I do not know the implications of using chmod function (PHP does not mention anything about problems in Windows). There was another error related to imported files size. It was also corrected (apparently).

Exporters hierachy is partially implemented. Bibtex, MODS and RIS output are ready (not tested yet). RSS will not be implemented according the published design because I want to take advantage of Joomla's RSS support. RDF is my new dilema so I need your help here. As you know, RDF is a general format for representing knowledge. I have decided to include RDF support in JResearch using XML flavour for serialization. The problem is that there nearly 100 ontologies (defined vocabularies for a set of entities) for representing publications knowledge . It is simple: I do not know which to use. I know some people is working with RDF in GSoC 2008, so I would appreciate a lot their help in that part. I will continue surfing the net those days to tidy up my ideas and find a solution. I suspect my job colleagues will also be helpful as they work in Artificial Intelligence.

Finally, my priorities:

•  Update the repository (URGENT)
• Automate the installation of plugins for persistent cited records.
• MLA citation style
• Integrate implemented classes to include exporter functionality (at least for implemented formats)
• Solve the RDF dilemma.
• JResearch backend ready by August 4th.

As always, comments and suggestions are welcome. Thanks in advance!!