• Why Allthingy?
  • Ye Olde Family Crest – Part I
  • Ye Olde Family Crest – Part II

Mike Urmeneta, Ed.D.

I Stopped Using Productivity Apps. I Built a Personal Operating System Instead.

by Michael

I want to tell you what happened the last time I walked out of a back-to-back meeting block with three things spinning in my head.

I took a screenshot of an email with a deadline in it. I sent a three-sentence voice memo to myself, half-formed, about a project that had stalled. I forwarded a registration confirmation I needed to remember. Then I went to my next meeting.

I did not open Notion. I did not update a board. I did not drag a card to a new column or type anything into a task field. I did exactly nothing organizational.

By the time I was back at my desk, everything was in the right place. The deadline was on my radar at the correct time horizon. The stalled project had a flag on it. The registration was on my calendar. My system had processed all of it, from a screenshot, a voice memo, and a forwarded email, without me doing anything beyond sending them.

That is what I mean when I say I built a personal operating system.


The problem with every tool I’ve tried

I have used most of the tools. Notion. Monday.com. Asana. ClickUp. Obsidian. Roam Research. I have built elaborate systems in all of them, and I have abandoned elaborate systems in all of them, not because the tools were bad, but because they all share the same underlying assumption.

The assumption is this: you are the operating system.

You configure the fields. You decide the categories. You update the statuses. You drag the card. You remember to open the app. You maintain the structure. The tool holds whatever you put in it, exactly as you put it in, until you come back and change it. It waits. It is always waiting.

This is not a design flaw. It is a design philosophy. Every one of those tools was built around the idea that the value is in the structure, and that you supply the structure. The better you are at maintaining the system, the more the system helps you.

The problem is that life doesn’t arrive in structured form. A deadline comes via a forwarded email at 6pm. A project update comes from a hallway conversation. An idea comes while you’re running. The gap between how life arrives and how these tools need to receive it is where every productivity system I have ever built eventually broke down. I didn’t lose discipline. The friction was real, and friction wins over time.


What’s actually different

My personal operating system is built on a different assumption.

The AI is the operating layer. I provide direction, judgment, and raw material. The system does everything else.

Before any session starts, it reads context: who I am, what I’m working on, what’s due, what’s stalled, what time horizon matters right now. It arrives with situational awareness. I don’t re-explain myself. I don’t reconstruct the picture. It already has it.

When input comes in, a screenshot, a voice note, a forwarded email, a few sentences I typed while standing in line, it parses the content, infers the structure, and puts things where they belong. Structure is applied after the fact, not before. I capture in whatever form the moment allows. The system handles the organization.

Every other tool requires you to be ready to organize before you capture. Mine requires only that you communicate.


What “frictionless” actually means in practice

Frictionless gets thrown around a lot. What I mean by it is specific.

I can send a photo of a handwritten note from a meeting and have it turned into action items. I can paste the body of an email thread and ask the system to extract the decisions and next steps and add them to the relevant project. I can describe something in half a sentence, “conference application, waiting on a letter from my dean, deadline first week of June,” and the system parses it, finds the right project, updates the status, and flags it if there’s no clear owner. I can take a screenshot of a social post with a deadline and have it converted into a calendar event without specifying the date, time, or format myself.

None of this requires me to already be in an organized headspace. It works precisely when I’m not.

The other tools require organization to receive input. Mine requires only that I communicate, in whatever form that takes on a given day.


Projects and tasks, differently

The thing I am most often asked about is how projects and tasks work. Because this is where most systems either become too complex to maintain or too simple to be useful.

My system runs on a radar rather than a task list. A task list is a queue. You move things through it. It grows, it shrinks, and the metric of success is whether it’s getting shorter. A radar is a situational awareness tool. Its job is not to be emptied. It’s to make sure nothing important is invisible when you need to see it.

Every project in the system sits at a time horizon: today, this week, this month, this quarter, or parked. That tag determines when the project surfaces. A project due this quarter shouldn’t be in my face today. A project with an action due tomorrow should be front and center. The radar knows the difference.

The morning review, which the system generates before I’ve had my first cup of coffee, surfaces only what’s relevant to the current horizon and flags anything that’s been stalled longer than it should be. The system reads the project list, notices that something hasn’t moved in twelve days, and tells me. I didn’t program that rule in any explicit way. I just described what I wanted, and the system enforces it. That is not how Asana works. Asana shows me a stale card. My system asks me what’s blocking.

For bigger projects, the ones that need more than a status and a due date, individual project documents hold the full picture: background, stakeholders, decisions made, open questions, next actions. These feed the radar but carry the depth. Nothing falls through the cracks because there’s no gap between the radar and the project detail. They’re connected.

There’s also a completed log. When something finishes, it doesn’t disappear. It moves to a record of what was done, when, and what came from it. Over time that log becomes a picture of what you actually accomplished, which is different from what you planned to accomplish, which is where most performance reviews fall apart.


The AI reads before it acts

Notion AI is smart when you’re already in Notion, already in the right document, already looking at the right thing. It answers questions about what’s already stored. At its best, it’s a very good search bar with generative capabilities. It does not know who you are before you open it. It does not know what you’re working on. It waits for you to provide the context.

My system reads context files before generating anything. There are documents that describe who I am professionally, what I’m trying to accomplish over the next quarter, what my working style is, how I want to be pushed and where I need to be held accountable. These load before any output is produced. The system doesn’t ask me to re-explain my situation. It already has it.

When I ask it to draft something, an email, a proposal, a reflection, it is not drafting from a blank slate. It’s drafting from a complete picture of who I am and what matters. The output sounds like me because the system has been trained on what sounds like me, not through fine-tuning or model training, but through documented examples it reads and matches every time.

Motion is the smartest scheduling tool I’ve encountered. It genuinely auto-schedules and re-optimizes as your week shifts. But Motion only knows your calendar and your tasks. It doesn’t know anything else about you. It solves the scheduling problem and leaves everything else untouched.

Obsidian and Roam Research, which I have real respect for, are built for people who want to own the architecture of their own knowledge. They’re intellectually serious tools. They’re also pure storage. You put in; you get out. Nothing flags. Nothing improves.

My system does both of those things. Every morning review generates a self-assessment of the template it used, with one specific improvement proposed. The system watches itself run and surfaces its own gaps. No tool I’ve ever used did that.


What I use to manage my whole life

Professional work is only part of it.

The same system that tracks a grant application tracks a recurring grocery run. The same system that drafts a strategy memo knows the payment schedule for a long-term financial commitment. The same system that surfaces a stalled work project notices that a family thing I meant to do three weeks ago hasn’t moved.

Every other tool I’ve used was built for a domain. Monday.com is a work tool. Asana is a work tool. Even Notion, which can technically hold anything, defaults to work structures and leaves personal life as a folder you never quite maintain.

My system holds all of it. Professional, personal, family, financial, all in the same structure, reviewed on the same cadence, surfaced by the same radar. There is no separate system for work and life because there is no separate work and life. There is just what needs attention.


Everything is markdown files

Most people don’t expect this part.

The whole thing runs on plain text files.

Not a proprietary database. Not a SaaS platform. Not a subscription that can change its pricing, sunset a feature, or lose my data in a migration. Plain text files, specifically Markdown, stored in a folder on my computer, backed up automatically, readable by any application that has ever existed and any application that will ever exist.

The intelligence isn’t locked in the files. The files are just context. The intelligence is in the AI that reads them. And because the files are plain text, any session, any device, any location, can pick them up cold and have full situational awareness within seconds. The system is portable because the files are simple. Any session can pick them up without syncing or re-orientation.

When I tell people this, the first reaction is usually: “That’s it?” Yes. A folder of text files and an AI that reads them before it acts.

The most powerful organizational system I have ever built is also the most technically humble. It doesn’t need complex software. It doesn’t need to be learned. It needs to be fed, with whatever you have, in whatever form you have it, and it does the rest.


What changed

Before this, I had systems I maintained. I was the operating system. I decided what went where. I updated the statuses. I noticed when things were stalled. I generated the weekly review in my head on Sunday night and hoped I hadn’t missed anything.

Now I have a system that maintains itself, and asks me the right questions when it can’t.

The morning review is already generated when I wake up. The stalled project is already flagged. The deadline is already on the radar at the right horizon. The email I forwarded at 6pm is already structured and filed.

I spend less time maintaining the system than I ever have. I spend more time doing the work.

That is what a personal operating system is supposed to do.

Posted in Uncategorized |
« The Expert in the Room

Leave a comment Cancel reply

Your email address will not be published. Required fields are marked *

About

Dr. Mike Urmeneta is an award-winning researcher, educator, data scientist, and storyteller with a passion for helping institutions improve and succeed through agile and collaborative approaches to research and analysis. He has extensive experience working with universities and has been recognized for his work by leading organizations such as the Association for Institutional Research, the National Association of College and University Business Officers, and EDUCAUSE. His diverse background and ability to build strong relationships with a variety of stakeholders have allowed him to make a significant impact on institutional policies, procedures, and priorities. Most recently, as an instructor for AIR’s Data Literacy Institute, he has been preparing leadership teams to embrace a culture of data-informed decision-making. Prior to this role, he served as the director of analytics and business intelligence for the New York Institute of Technology, providing strategic guidance to various departments, the president’s office, and the board of trustees. Dr. Urmeneta has also held various administrative roles at New York University, including in admissions, financial aid, enrollment and retention, alumni relations, and development. He holds a Bachelor of Science in Mechanical Engineering and a Master of Science in Management from NYU, and a Doctor of Education degree from Northeastern University, where he received the Dean’s Medal for Outstanding Doctoral Work for his research on first-generation college students.

Categories

  • AI (1)
  • Analytics (15)
  • ChatGPT (1)
  • Doctorate (16)
  • Education (25)
  • HigherEd (1)
  • Infographic (25)
  • Leadership (13)
  • Personal Development (24)
  • Professional Development (21)
  • Python (7)
  • Tips and Tricks (14)
  • Tutorials (5)

CyberChimps WordPress Themes

© AllThingy