As a mental exercise to keep my brain active and thinking, and my IT skills from “ossifying”, I am trying to “wrangle” the mathematical model for what is called the “Mexican Method” which was shared with the public via the Geodesic Help Group on Google Groups by Mr Héctor Alfredo Hernández Hernández, currently with the Universidad de Sonora, Mexico.
It is a technique for optimizing the vertex placement for a Geodesic grid, in order to make the process of building those more economically viable, by minimizing the number of customized “cuts” required for the various components (a.k.a. minimizing the number of different strut lengths). As structures get larger (referred to higher frequency for complexity of grid), the variability in struts becomes unmanageable, without such optimization. The Mexican Method does so by
first identifying vertices that are suitable for applying placement optimization, then
performing a vertex displacement, either tangential to the “inscribed sphere” or, in some cases, by radial displacement off the sphere.
Members of the Google Group have shared spreadsheets outlining the results of their optimization, but these never seem to include the code that performs the actual optimization itself.
My previous attempt was Python-based, but that was too foreign a mind-set for me to overcome, and I only got as far as generating the grid array, before abandoning that first attempt. Currently working at this via use of the C-Language.
For now, I’ve put a pause on my efforts to create my own instance of what I refer to as “Yet Another Battleship”, which is a project to create an implementation that is intended to be standalone on a User Desktop, using HTML, CSS and JavaScript, and which is simple enough and contains all the embedded documentation to make it a “learning vehicle” for those trying to master those technologies. Some of the work I did on that was shared in the old Ubuntu MATE Forum.
I’m testing the (newish) Unifi firewall against OPNsense.
I’m currently working on post-install configuration for Wazuh:
I’ve got agents deployed on all servers and some workstations
Currently working on centralizing log mgmt
I recently deployed the Crowdsec console in docker compose and I’m working on configuring it as a LAPI.
I recently migrated all of the logging/metrics/tracing agents on my servers from multiple agents (node-exporter, promtail, etc) into Alloy, so I now have a single agent deployed that collects everything instead of multiple agents. However, even though Alloy has a built-in node-exporter feature/plugin, none of my metrics based Grafana dashboards work now. I can still see metric in Grafana Explore, so I need to dive down into that rabbit hole.
I’m also working on moving more tasks in ansible (a never-ending task). With one of the recent updates, all of my playbooks broke, so I still need to dive into that rabbit hole as well.
Earlier back, I saw mention of GL-iNet. I have a Berl AX that I use as a travel router. When I travel, I usually take a roku or two with me. With the travel router, once I get it connected all of my devices connect to it’s wifi, and I use the button the Berl AX to enable VPN. The convenience factor is that I don’t have to touch any other device, they simply connect to wifi and I’m online.
