Since I've been having severe issues with our project on eclipse, I have been looking into Intellij by the recommendation of Zach Harding. I really hope that it is as good as it claims to be and will resolve my issues.

The lecture from the guest speaker was really interesting to me. The main reason being that I see a lot of similarities of what he was talking about with my current internship. A lot of the processes sound the same, the commit sizes are small because of a mature code base, etc. The only difference is that there isn't too much significance of time at the moment, or it may just be me being out of the loop since I'm only a lowly intern.

Things are looking really good for our project. Dan's girlfriend has delivered on a massive scale with a wonderful series of mockup images. Dan is thinking that the backend is almost done, and that we can begin focusing on working on implementing the front end to look like the mockup images.

I've been having issues with my eclipse project folder settings for a while so I haven't been able to keep track of everyone's work; so I have just been working on the calendar portion and some of my own simple mock up pages that are coded to start off with.

Our project is moving along well. Dan recently got a remote server up and running using Amazon's AWS services. Our backend is looking to be in good shape and we've started on developing a basic front end. Unfortunately, I've been running into project/repository issues with the recent project consolidation/cleanup from the initial repository. Hopefully I can get that resolved quickly with the help of Dan.

I've found something very promising for the calendar visualization. There is an AngularJS framework based utility packaged called AngularUI. AngularUI has a full calendar (month view, week view, day view) that I think we'll be able to use for our project. I've started looking into it and hope to have a demo up within the next couple of days.

I've been tasked with researching how we are going to visualize the schedule. It will most likely be some sort of weekly calendar, similar to what UNM has for their weekly schedule or Google's calendar, in which you can add schedule objects that will then appear as blocks of time in the calendar.

I'm feeling good about our project, it looks like all the separate pieces that we've been working on are fit together well and we're able to get basic query functionality. With the basic guts implemented its now time to start working on the front end design of things and make it look pretty.

It has been a really interesting time figuring out how we're going to store and update our SQL database for the potential feature of notifying registered users (another potential feature) that a class that they chose for their schedule has changed. We've had to change routes a couple times already, and know that what we were doing wouldn't quite work correctly. We still have quite a bit to discuss to figure out exactly how we are going to implement our database structure for this feature. One thing that we do know is that it will most likely have some sort of global version control that we will then utilize for various things such as deleting old versions, and comparing old version data to the new version data to look for differences.

Recently,  I've been catching myself thinking about what type of algorithm we're going to be using to compute schedules. I've been trying to think of various ways to do it, but have not really thought too deeply; I've mainly just been accruing starting ideas to then run on and see where they might lead to when it comes to the point of actually developing the algorithm. I've tried searching online to see what might already be out there, and found a thing that used a genetic algorithm to create a class schedule. Unfortunately, we have many more variables that would have to be accounted for, and the genetic algorithm is probably out of scope for this project at the moment; the best bet for now is to create a simple naive algorithm or a greedy algorithm to create schedules when the time to comes.

Well, finally got around to writing a bash script to download the xml data from the opendata unm website. Juggling work and classes has been much more difficult than expected. The bash script is just a simple wget followed by the url to the download link. It can be run from anywhere, but there is no central server yet, thus there is no reason to start a cron job to keep the xml up-to-date while we are still in component development.