The end of Spring 2012.

Alright, with the semester rapidly coming to an end there's no better time for a wrap-up post. Overall this semester has been great, both in and out of TWC. Unlike most of the other undergrads in the TWC lab I tended to bounce between the major projects and work on smaller projects within them. In the beginning of the semester I continued to work on the BCO-DMO site and by the end I was working on the main Druple site for TWC. Some of the things that I did this semester include adding javascript buttons/arrows to the BCO-DMO site, re-crafting the generate.php function on the Druple site to make it more generalized and modular, and now I'm working on making a TWC Sparql Tester Site, maybe TWC STS will work as a title. When it's done, the STS will allow the user to select Sparql queries and XSL's, choose an endpoint, enter Sparql parameters, and then get back a preview of what the Sparql+XSL webpage would look like. Right now I have most of the user input code finished, in a mix of html/php/javascript, and we can hopefully use the TWC Sparql module to complete the query portion. I've had to stop working on it for now due to my finals schedule, which squeezes every final and project into the last two days of class, this coming Monday and Tuesday. Even when the semester is over, that doesn't mean that I'm done with TWC. I'll be staying in Troy during the summer in two part-time postitions, one here at TWC and another with an RPI department working on a research project about the disruptive effects of new technology on the consumer and industrial markets. I'm looking forward to the coming months and continuing with TWC come the Fall semester.

Post Undergrad Meeting

With a bit of testing we managed to finish all the work for the new generate.php and its associated files. The main generate.php file works perfectly, the files that xsl files (person, projects, organizations) are in working order, and some pointless/outdated files (meetings, authors, researchers) have been removed. some of the errors were coming from random characters misplaced in the files and others seemly came from no where and fixed themselves over the course of testing. With this out of the way I'm starting up on a new way to test queries, xml's and xsl's outside of the usual Drupal edit/view method. I'll write more once I have my plans set up and a clear picture of how I'm going to code it all together.

April Showers.

I haven't been able to test if  my work on recrafting the generate.php file for a few weeks since something went wrong with the Leo test site. Even though I'm still a bit unsure of what happened, it sounds like "someone" loaded the wrong files so that everything was looking in the wrong place and thus failing to properly load. Once it was back up we got to start testing and right away found some issues. The generate.php file that I had prepared was no where in sight and in its place was a no functioning copy of the old generate.php, it had duplicate lines of code that stopped it from actually doing anything. As I'm typing this, I'm trying to track down some of the error logs that will hopefully show why the project pages are failing to properly transform the xsl files. Once that's done everything should be in place and I can move onto the next project. I'm going to be splitting my time for the next week and a half on TWC work and prepping for my CCNA certification test, which I'm hoping to pass on the first try. And we finally get a TWC Undergrad meeting. Yay!

March Madness

After some midterms, spring break, and even more midterms March is just about over. Before leaving I finished setting up the new xslt files for the generate.php recrafting, but didn't have time to implement and test them. After I got back, I put everything into place and started to test out the new files. Unfortunately, the test machine started to have hiccups that made it difficult to tell where the generate function was working and where it wasn't. During one of the good days everything seemed to be working, but the next everything went south. As of now something appears to be wrong with most of the leo machine and only a few select pages still return anything. Hopefully one of us will find something in the one of the logs (system, php, apache, and escience) that will help fix the problem. Still no word on what I'll be doing for the summer, since the only organization to actually get back to me was Woods Hole, who unfortunately decided not to take me. For now I'm focusing on getting my paperwork in order to sign up for the co-terminal program her at RPI. I'm still undecided on who I want to ask to be my graduate adviser and what areas I want to focus on. Let's hope that I can push everything into place without too much trouble.

Getting the Ball Rolling - Spring 2012

Before the official email went out asking who wanted to come back to TWC in the Spring I was already dead set on continuing in the Undergrad Lab. The first thing I did this semester was put some more work into the BCO-DMO site. I added arrows for every drop-down menu to show if it was open or not. To stick with the original site's feel I used the same arrow images on the new version. I lost a week in productivity to a cold before I got started on my next bit of work. I'm now "recrafting" an existing generate.php file to make it less specific and more general. Looks like the first thing I'll need to do is get a hang of how the code works before I start going in and moving things around. Let's hope nothing goes too wrong. I'm currently looking into opportunities for summer employment/research and have already applied for several position. My current first choice is a student fellowship at Woods Hole Oceanographic Institute, but if I don't get accepted I'm looking to stay in Troy and do something on campus.

Blog. Blog.

Original Osprey Site

New Druple Triple Store Powered Site

Despite being a junior, this was my first time participating in an URP and my first introduction to the Semantic web.  I had heard the term ”semantic” once or twice, but never gave it much thought as anything more than being another study in computer science that I would never delve into. All that changed when I joined up with the Tetherless World Constellation.  I browsed through the open projects and eventually settled into the website recreation for the Biological and Chemical Oceanography Data Management Office, or BCO-DMO for short.

The original BCO-DMO site used a more traditional relational database, but the goal of TWC’s version was to show that an external triple store could be used to accomplish just as much, if not more. At the onset I had absolutely no idea how to use any of the tools, languages, or logic systems that would form the meat and bones of the new site. On the basest level we needed to learn and implement a language that would be able to draw data out of the triple store for future use. This of course meant that we needed to use the Sparql Protocol and RDF Query Language, or Sparql. Each aspect of the new site needed its own Sparql query to pull the needed data; each category of data had one query for returning a general list of every member of that category and another query for returning specific data for each of these individual members. Once we had a way to find the data we wanted, we needed something to use it, that’s where Xpath and XSLT came into play. By having the Sparql queries return their results in XML format we could have XSL files converting those results into HTML that can easily be used on the website. I felt like this was a biggest bottleneck since our entire team seemed to grind to a halt when it came to actually writing XSLT code. It almost seemed like we weren’t going to finish this semester until Ali Nendick joined onto the BCO-DMO team. After I saw her making more progress in a day then we made in weeks I was compelled to try and keep up. Although I still can’t keep up with her dedication and speed, I feel that I have gasped all the basics that make up the project and could reproduce my work a lot faster and more efficiently. Overall, the project is somewhat coming to an end; just a few more pages need to be tweaked and then everything could use a proverbial coat of paint just for show.

After having a great time this semester, I’m looking forward to continuing my work with TWC in the future. Now that I’m over most of the awkward having-no-idea-what-to-do phase I think that I can contribute more and more to the project I’m involved on.

Work. Work.

I never feel like anything thing that I've been doing is worth putting up on this, but might as well write about something.
Since Ali started on the BCO-DMO project it seems like there's been a lot more progress then the entire rest of the semester, which is fantastic since I was spinning my wheels and I had no idea what the rest of the group was doing. The site seems to be coming together into a rough, but working, form. I spent a few hours last week trying to get javascript to work within the xslt so that we could make things look a bit nicer and add the collapsible tables to the individual sections, but I never did get it to work properly.
Let's hope this all gets working soon.