About Odesys · since 2009
17 years and counting.
Back in 2009, it was a different time. I was studying at FOV, the web was still young and I needed a project for my thesis. There was a subject close to my heart called Decision Support Systems. Professor dr. Vladislav Rajkovič captured us with his charisma and stories.
-
2009
The spark
At one point, we were learning how to plot a decision model on the ABACON chart. I found it engaging to be able to visualize your decision model, but pen and paper? Surely, we can do better.
So I combined my amateur web-development skills with a young man’s passion and half-thought out ideas and produced something remarkably terrible.
-
2010
The thesis
By July 2010 the terrible something had a name, a method and a diploma thesis behind it: ODESYS — the ABACON method, adapted with a few Kepner-Tregoe ideas and wrapped in a five-step web flow. The repository’s first commit is dated May 21, 2010, and it has been the same repository ever since. Having your own project during studies turned out to be a smart idea, too — I sneaked it into every course I could. Class about Computer Security? Perfect opportunity to learn about SSL and securing a server.
-
2010–2013
Frenk makes it real
Five months after the first commit, my good buddy Frenk jumped into the codebase — his first commits are design fixes — and he never really left: nearly five hundred commits over the next four years. In 2013 he went all the way and rebuilt ODESYS as his own master’s thesis: the modern, cross-platform version on the Yii framework, with the group-decision idea baked in, validated in a live experiment with students at FOV. The ODESYS that students used for the next thirteen years is as much his as it is mine.
Frenk’s 2013 thesis (PDF, Slovenian) · my own 2013 thesis (PDF, Slovenian) — mine was about multiplayer web games, but Frenk was on that team too. Some collaborations just keep going.
-
2023
Out of the closet
In 2023, I took it out of the closet, dusted the server and considered the rewrite it deserved — and decided I couldn’t justify it. Even with early AI tools helping, a full rewrite cost more than a hobby project for a few hundred students could carry. So: fix what’s broken, throw on some paint. Mobile got usable, the design got fresher, and Yii and jQuery stayed right where Frenk left them.
And you know what, it was fun to look at our code from 10+ years ago. I even left the camp a bit better than when I found it.
-
2026
The rewrite, finally affordable
In July 2026 the rewrite I couldn’t justify three years earlier took seven days. The Yii app became a Laravel app — ported behind a characterization test suite committed before the old source was deleted, so the behavior Frenk and I shipped as students survived, verified, into the new engine. What changed wasn’t ambition; the way I build software did.
The group features Frenk explored in his thesis finally grew into the product — you can now ask a group before you decide. Still mostly useless. Still our playground.
Yet somehow, after all these years, any time there is downtime, I get an email. Students are still using it to learn about decision support systems.
