Recent Blog Posts


Website Update
April 22, 2023

Welcome to the new www.DavidABeeler.com! I have been meaning to update this website and expand its capability to be more useful and meaningful. From 2019-2023, the website was built with GatsbyJS and hosted on Netlify. Interestingly enough, Netlify acquired GatsbyJS to "accelerate adoption of composable web architecture," which continues the (positive) march towards microservices. Typically, behemoth architectures are undesirable; due to the extreme interconnectivity of all pieces, they are harder to develop, enhance, and maintain. Towards the end of 2022, I started working on a portfolio website for my niece, who is graduating from high school this year … read more

Recent Portfolio Pieces


Erma Bombeck TV Scene
February 13, 2024

As a small portion of the "Beyond the Byline: Erma Bombeck's Story" exhibit in the University of Dayton Roesch Library, I put together a video that runs on a loop in the space (you can see the video in the intro of the exhibit opening YouTube stream). This project was great for expanding my knowledge of DaVinci Resolve and finding a presentation solution for low-resolution clips.

Based on our first test with the low-resolution clips, displaying them full screen on the TV was not ideal. I had the idea to incorporate a retro TV scene to keep the videos smaller but still visually interesting. I created a composite image in GIMP (GNU Image Manipulation Program), which was a successful start, but it was a challenge to find freely available photos that checked all of the boxes.

To have control over the scene, I decided to recreate it in Blender. Using a few reference images for a retro-style television and Erma's typewriter, I modeled everything in the scene except for the flower and vase. Through trial and error, I found lighting, HDRI, and rendering settings that produced a reasonably realistic result. The clips in that portion of the video were on the air in different years, so I created individual renders updating the "On Air" sign appropriately.

This was a ton of fun to put together, and I am pretty happy with the final result. After seeing all of the photos in the exhibit, if I were doing it over, I would replace the image above the TV with something else. I also would have modeled something to replace the flower, possibly putting in a few of Erma's books, centering the TV a little more.

Library of Congress Call Number Game/Training
July 26, 2023
Demonstration of the Call Number game/training showing multiple books on a bookshelf where you must rearrange the books to the correct Library of Congress order.

A colleague has been working to streamline student worker training for Access Services in the University of Dayton Roesch Library. Initially, they sent links to Kent State University and Carnegie Mellon University's training to see if we could do something similar. In Kent State's case, their JavaScript code was homegrown and used SortableJS and CMU built their game using Flash and ruffle.

That following weekend, I put together this little MVP - try it out! The initial version was designed for a large, high-resolution monitor, so it is not a great experience on mobile. Improvements are incoming as we will link to this training/game from Isidore and use it this fall (2023).

This is a JavaScript solution using the Dragula library for the drag 'n drop functionality, a JSON structure for easy modification of the tasks, and template literals to build the books and shelves. At the moment, there is no internal logic to determine the correct order; order is specified by the JSON structure.

Upon completion of the training, the user can input their name and the jsPDF library will generate a basic certificate with a pixel aesthetic. This can be uploaded as part of the training process to confirm the training was completed.

Repo: https://gitlab.com/dabeeler/call_number_training_game

This project was inspired by and took some design cues from the following projects:

UD Rose Gallery VR Experience
March 20, 2023

As an example of the technology and a peek of what I could create within a short time frame, I set out to incorporate the DeGrazia Madonna, a variety of Mother Mary statues, and the Erma Bombeck Rose Gallery exhibit from 2022 into a virtual reality (VR) experience on the Quest. As a simple art exhibit, I added UI panels for each of the statues and wall clings to describe the piece and recorded voice-over audio. The Rose Gallery was created in Blender, the wall clings were photographed and turned into usable textures, the DeGrazia Madonna was turned into a 3D model using Meshroom, and "The Juggler of Notre Dame" stained glass was modeled in Blender and textured painted from a photograph. The wind/water mill was modeled in Blender based on a course. These assets and models found on Sketchfab were used to create a simple environment in Unreal Engine 5.

Attributions:

Juggler of Our Lady Exhibit Technology
November 03, 2022
Technology provided for the Juggler of Our Lady exhibit at the University of Dayton Roesch Library. This included a television with a looping slideshow and an iPad with an interactive slideshow.

I had the opportunity to assist with the "Juggler of Our Lady" exhibit at the University of Dayton Roesch Library by incorporating two pieces of technology. One aspect of this was a television displaying a looping slideshow to provide context to and display an animation. Near this television, we mounted an iPad to an old lectern to display hand-drawn storyboard panels of a different animation of the same story. Although both displays used RevealJS, a wonderfully easy-to-use presentation framework, the iPad also incorporated a pan-zoom library so the attendee could interact with the panels. I also developed a basic engagement tracker for the iPad display, recording sessions. These displays were easy to integrate into the exhibit and can be easily modified for future exhibits.

Statue to 3D Model
June 08, 2022

The Marian Library at the University of Dayton has various Mary-related statues. This DeGrazia Madonna statue was previously photographed using a rotary table. I used Meshroom, a photogrammetry application, to generate a 3D model based on these images. Although the photos were not ideal for reconstruction, Meshroom created a reasonably clean model. I used Blender to smooth out inaccurate geometry at the head and base and corrected some areas of the texture. The resulting model is imperfect, but it was a successful test of capability based on previously captured photos and free software. See the model at Sketchfab.