Obsidian as a Long-Term Memory for Multi-Year Research Projects

My PhD changed topics, supervisors, and directions — my notes survived all of it. That’s when I realized Obsidian wasn’t a note app for me, but long-term memory for a multi-year research project.

When I started my PhD, I genuinely believed I would “basically remember” what I worked on. Maybe not details, but at least the big ideas. After all, I was deeply immersed, motivated, and constantly thinking about my topic. How could I forget something I spent months on?

Fast forward a few years, and I can confidently say: that belief was adorable.

PhDs are not linear projects. They are slow, chaotic, multi-year negotiations between ideas, people, methods, and reality. Topics drift. Assumptions collapse. Supervisors change. Side projects die. And your own thinking evolves in ways that make your past self feel like a stranger.

Most note-taking systems are not built for that.

The uncomfortable truth: folders assume stability

My early setup was neat. Painfully neat.

Folders for:

• Topic A
• Topic B
• Methods
• Experiments
• Drafts

It looked professional. It looked controlled. It also quietly assumed something that turned out to be false: that my project structure would stay stable.

But research doesn’t work that way.

What happens when Topic A becomes irrelevant? What happens when a method moves from “core” to “background”? What happens when a dead side project suddenly becomes relevant again two years later?

Folders don’t answer these questions. They freeze past decisions in place.

Why long projects break traditional systems

In a long PhD, you don’t just forget content. You forget context.

You forget:

• why you dismissed an idea
• why a method felt promising at the time
• why a paper annoyed you
• why something didn’t work

And this is dangerous, because future-you will confidently repeat old mistakes if past-you left no trace.

Word documents, folder hierarchies, and “final” notes are optimized for completion. A PhD needs something optimized for survival.

Obsidian as memory, not storage

The shift for me happened when I stopped treating notes as storage and started treating them as externalized memory.

Memory is not a list of files. It’s a network:

• ideas connect to ideas
• failures connect to assumptions
• methods connect to constraints

Obsidian doesn’t force you to model this perfectly. It just allows it to exist.

Links don’t care whether an idea is “old”, “dead”, or “out of scope”. They just preserve relationships.

That turns out to be everything.

Notes that age well (and notes that don’t)

Some notes age terribly.

These are the ones that assume permanence:

• polished summaries
• “final” conclusions
• rigid structures

They rot the moment your thinking changes.

The notes that survive are the ones that capture:

• reasoning
• uncertainty
• decisions

Instead of writing “This approach is bad”, I started writing:

“Rejected because it assumes X, which we later realized doesn’t hold for Y.”

Years later, that note is gold.

Not because the decision was correct — but because the reasoning is still legible.

Dead projects are not dead knowledge

One of the most underrated benefits of Obsidian shows up when projects fail.

In a folder-based system, failed projects get archived, zipped, or ignored. They become intellectual landfill.

In a link-based system, dead projects quietly dissolve into the network.

A method you tried for Project A might resurface in Project C. An assumption that broke Project B might save you months later. An idea that felt premature suddenly becomes obvious.

Because the notes were never locked into a project folder, they remain reachable.

This is incredibly important in multi-year research, where relevance changes faster than you expect.

Topic changes without data loss

At some point, my PhD shifted direction. Not dramatically, but enough to make old structures useless.

If my notes had been organized strictly by topic, this would have been painful. Instead, most of my notes didn’t “belong” to a topic at all. They belonged to ideas, methods, or questions.

So when the topic shifted, I didn’t reorganize everything. I just started linking differently.

Old notes didn’t disappear. They just became less central — until some of them weren’t.

That’s exactly how memory works, by the way.

Long-term clarity beats short-term neatness

Obsidian notes often look messy if you inspect them individually. That bothered me at first.

But multi-year clarity does not come from neat notes. It comes from traceability.

I want to know:

• where an idea came from
• how it evolved
• why I changed my mind

That story is far more valuable than a clean summary written for a version of me that no longer exists.

Writing in year four with notes from year one

The real test came late in my PhD.

I was writing sections that depended on ideas I explored years earlier. I barely remembered them emotionally — but the notes were still there, connected, readable, and honest.

I could see:

• my early enthusiasm
• my naive assumptions
• the moment things broke
• the point where I moved on

That continuity is impossible to reconstruct from polished documents alone.

It only works if you allow notes to be incomplete, contextual, and connected.

This is why Obsidian beats “final documents”

Final documents lie.

They hide uncertainty. They erase failed paths. They present decisions without the struggle that led to them.

That’s great for publications. It’s terrible for thinking.

Obsidian lets you keep the mess — and that mess is what carries meaning across years.

A PhD is longer than your discipline’s trends

Methods go in and out of fashion. Terminology changes. Reviewers suddenly care about things nobody mentioned before.

If your notes are tied too tightly to current terminology or structure, they age badly.

Link-based notes age better because they are semantic, not cosmetic.

The names can change. The relationships remain.

What this system protects you from

Most importantly, it protects you from your own amnesia.

It assumes you will forget. It assumes you will change your mind. It assumes future-you will be confused.

That is not pessimism. That is realism.

Final reflection

I don’t think of Obsidian as a productivity tool anymore.

It’s closer to a research prosthetic — something that compensates for the fact that a human brain is not designed to hold five years of evolving, technical, emotionally charged work.

My PhD is still messy. My thinking still changes. I still forget things.

But the memory is there. External. Patient. Connected.

And when future-me inevitably thinks, “Why did I ever believe this was a good idea?” — the answer is usually just a few links away.

If you’re early in your PhD, this might sound abstract. If you’re deep into it, you probably already feel the pain. Either way, build for the long term. Your future self will not remember — and that’s exactly why your notes must.