Nathan Capstone Blog

I suppose my greatest success was finally figuring out how to get to use GitHub to have code I was working on locally get committed and pushed on to GitHub, and how to pull and merge my branch with the main branch. Because it’s been more than two years since I used to GitHub, and even at that time it was not done in a collaborative setting, so when starting this project I really did not know what I was doing, and when I tried to make a pull request I would feed back from teammates saying how my code was not up to date, and how I had erased some of their code, and there were merge conflicts. So finally, after not doing it right for so long, I was able keep my code up to date and resolve merge conflicts. This may sound trivial, but I think reacquainting myself with the tools was probably quite useful.

As far as SWOT goes, I suppose the strength of the course is that it introduces you to working in a collaborative environment, and the tools that are necessary for that, which was not something that I otherwise had. Furthermore, the project may have served as a good opportunity to learn something that was not really addressed in the course. I also thought that the course was good in that in encouraged students to do things to prepare for finding a job. Also, because of the collaborative nature of the project, there was some networking opportunities. I guess the weakness was the short amount of time we had to do it. Otherwise there are not really any threats I could come up with. From a purely personal perspective, I think it would have been good if perhaps there were more projects related to systems programming, but that is just a personal preference.

My favorite project related technology? Difficult to say. I feel our project is relatively low tech. It mostly seems to be a lot of hard coding of things. That said I suppose that my favorite bit of tech would probably be the graphical interface, for which we are using the tkinter Python library. I have used once before, when I had no prior GUI making, and I can affirm that is a very intuitive and convenient, and user friendly software to use. I highly recommend it. The GUI, for me is the best part of the program since it actually produces something that the viewer can see and make sense of, and feel like they are witnessing something evolving right before their very eyes. Basically it’s the technology that makes the simulation come alive. Otherwise, you are just left with a bunch of data.

I can’t say we are using AI a whole lot. I am occasionally using ChatGPT for help, which is really quite useful. But ultimately it just ends up being an advance and specific search engine. In that way it is quite convenient, somewhat like having a personal teacher. Otherwise, I guess I would like to implement something in the project that might somewhat be considered an AI. I would like to have organism that when faced with a decision, make the decision randomly with a given probability, and then based on the outcome of that decision, have that probability be modified. For instance, if a decision leads to an increase in energy or in successful procreation, the probability of doing that might be increased. Hopefully, overtime, this would lead to a greater prevalence in that advantageous behavior. I had heard similar strategy was used in early AI models in the 50s, in tic-tac-toe, where a move from a certain position that lead to a win, would then have its probability increased. Overtime, this lead the program to play better games. However, this feature has yet to be successfully implemented.

It seems that I am now approaching the end of my time at OSU as this class will be my last. I will also be separating from the Air Force, where I did weather, at around the same that this class ends, so I am really in a transitionary state. Originally I was planning on doing a degree in meteorology as there was program at the University of Arizona that liasoned through the Air Force to offer enlisted USAF weather personnel a BS in meteorology. Unfortunately it seems that this was discontinued so I opted into a post-bacc program at OSU instead. I have no particularly deep thoughts on the matter, my only hope is that it all goes as smoothly as possible I can transition into good job. I am currently stationed in Omaha, Nebraska. Afterwards, I will likely live in France, where my parents are, and try to find a job there.

In choosing a project I was rather torn between A-Life Challenge, which seemed to me to be the most interesting, and doing the ML Breakout project which seemed to me would be the most useful in terms of development since it purported to be a good introduction to AI, which is an area in which I feel it would be useful to gain experience. It was a very difficult choice, since I do find the idea of a simulated evolutionary environment extremely compelling, however, ultimately, I ended up ranking the ML Breakout project over the A-Life Challenge, because I felt the practical skills I could learn could help further learning later on. That said, if I end up with A-Life Challenge, instead I won’t exactly be disappointed. Alternatively, I list the Emulator project as 3rd as I do have an interest in the lower level aspects of computing and that seems to me to be a good project to learn that kind of stuff. I had originally considered ranking it first. However, I realized most of the materials to do similar projects were readily available online, and so I figured I could the project some other time.

Welcome to blogs.oregonstate.edu. This is your first post. Edit or delete it, then start blogging!