ModuleCycle

Part I — WebDAV, MinIO, rsync, or Syncthing for VPS File Sync

Tue, 07 Apr 2026 00:00:00 GMT

1. Introduction

1.1 Why I needed sync between a VPS and my local device

I often find myself living in at least two different worlds. One is the place where I read, test, deploy, and solve problems on my VPS. The other is my local space, where I think more slowly, write notes, and try to turn scattered ideas into something meaningful. Knowledge usually appears in small sparks across these different scenes. If I do not catch them in time, they fade very quickly.

In the end, I chose Obsidian as a bridge between these fragments. It gives me one place to collect commands, mistakes, ideas, and small conclusions. But collecting is not enough. If I do not review what I wrote, I forget it surprisingly fast. That is why syncing matters to me. I want these notes to move with me, from server to laptop, from one device to another, so I can return to them whenever I need.

There is also a deeper reason. I have a simple dream: to spend more time linking my knowledge together. Not just storing it, but connecting it, refining it, and making it available anywhere and anytime. A good sync tool is not only about convenience. For me, it is part of building a personal knowledge system that I can truly rely on.

1.2 The real problem: not just “can it sync”, but “which tool fits my workflow”

At first, I used Nextcloud WebDAV to manage my files. For a while, it seemed good enough, especially because I relied on the Obsidian Remote Save plugin to keep my notes connected across devices. It looked like I had already found the bridge I wanted.

But after using it for some time, the real problems started to appear. When I needed to sync a large number of files between two devices, WebDAV often became slow and frustrating. Nextcloud itself also brought extra pressure: frequent updates, ongoing maintenance, and the feeling that I was spending too much time taking care of the tool instead of using it.

That was the moment when I realized the real question was not simply whether a tool could sync files. Many tools can do that. The more important question was whether a tool truly fit my workflow. I did not just want something that worked in theory. I wanted something light, stable, and easy to maintain over time.

So I asked AI for ideas. It suggested several other options, including S3-compatible storage and Google Drive. Those alternatives immediately caught my attention. They felt different from the traditional WebDAV path I had already known, and that curiosity pushed me to start exploring a wider range of solutions.

1.3 My two main directions:

After that, I found myself moving in two different directions.

The first direction was still centered on the file-service model I already knew. Instead of continuing with Nextcloud, I started looking for a lighter and more stable replacement for WebDAV. At the same time, I also began to consider whether I should step away from WebDAV itself and try more direct tools such as SFTP, rsync, or rclone.

The second direction was quite different. Rather than thinking in terms of folders, mounting, and file services, I began looking at object-storage-style solutions. This was a new path for me, and it felt more modern, especially because many sync tools seem to work better with S3-compatible storage.

Based on AI’s recommendations, MinIO quickly stood out. It looked like an interesting bridge between self-hosting and the S3 ecosystem, so I knew I would want to try it soon.

1.4 Temporary conclusion

At this stage, I have not reached a final answer yet, but the direction is becoming clearer. What began as a simple need for syncing notes between my VPS and local device gradually turned into a broader reflection on workflow, maintenance cost, and long-term reliability. I realized that the problem is not just about finding a tool that can move files, but about choosing one that fits the way I actually work.

For now, I have narrowed my exploration to two main paths: traditional file-based solutions such as WebDAV, SFTP, rsync, or rclone, and object-storage-based solutions such as S3-compatible services, especially MinIO. Each path seems to represent a different philosophy of managing data, and I want to understand not only how they work, but also how they feel in daily use.

I do not have the full conclusion yet, but at least the question is now more precise than before. That already feels like progress. Tomorrow, I will continue by testing these options more carefully and comparing which one is truly the best fit for my personal knowledge workflow.

[DevTools] Modular Knowledge Management with Git Sparse-Checkout and Obsidian Git Integration

Mon, 15 Sep 2025 00:00:00 GMT

🏗️Background

We plan to establish three different types of databases to handle knowledge environments in different scenarios:

The standard database, named standard-db, is used to record and revise historical experiences and knowledge conclusions.

The creative database, named creative-db, is used for creating standardized NotionNext documents such as blogs and tutorials.

The workspace database, named workspace-db, is used for managing work tasks and knowledge learning.

🧱Main Content

🔄 Building a Seamless Knowledge Workflow: From Obsidian to GitHub to Vercel

Let's first focus on maintaining the standard database — how to elegantly manage historical experiences in Obsidian, perform git pull, and ultimately use GitHub Actions to push updates to Vercel, enabling automatic backup and updates of Hugo. The workflow is illustrated in the Mermaid flowchart below:

By exploring the Obsidian plugin library, we identified a plugin that fits our needs: Obsidian Git. After reviewing its 📖 full documentation, we confirmed that it supports all the essential features required for our workflow:

git pull: Supports both automatic and manual pulling of remote updates.

git push: Allows automatic or manual pushing of local changes to the remote repository.

git commit: Supports automatic or manual commits of changes.

Conflict detection and handling: The plugin does not auto-merge in the event of a conflict. Instead, it notifies the user to resolve conflicts manually.

Auto-sync intervals: You can configure it to automatically pull/push at defined time intervals.

Sync on startup: Automatically performs a git pull when Obsidian is launched.

Command palette integration: All Git operations can be executed via Obsidian’s command palette, eliminating the need to use a terminal.

This plugin fully meets our basic requirements for maintaining the standard-db within Obsidian while seamlessly integrating with GitHub.

🧩 What Are Git Submodules Used For?

Git submodules allow you to nest one Git repository inside another, which is especially useful for:

Referencing external dependencies (such as themes or libraries)

Managing dependency versions independently

Avoiding copy-pasting third-party code into your main repository

In Hugo projects, submodules are commonly used to include themes. For example:

This command automatically updates the .gitmodules file and adds the theme as a submodule under version control.

If you decide not to manage blog/themes/meme as a submodule, you can simply delete the .gitmodules file and remove the submodule itself from your repository.

🧨Case: Why Sparse Checkout Didn’t Work as Expected

🎯Scenario

I ran the following commands:

I expected to fetch only the content/ directory, but I still saw many other files in the root directory:

🪤Root Cause

By default, Git uses cone mode for sparse-checkout. In this mode:

It only supports directory paths

It does not exclude root-level files, even if I didn’t specify them

So even though I only set content/, Git still checked out all the files in the root directory.

✅Correct Approach

To fetch only the content/ directory and ignore everything else, I needed to switch to non-cone mode:

This way, only the content/ directory is checked out.

🧪 Final Fix (Step-by-Step Commands)

To fix the issue and fetch only the desired directory, you can copy and paste the following:

🛠️ Bonus: Fetch Multiple Directories

If you want to include more than one directory (e.g., content/ and docs/), update the sparse-checkout file like this:

This will fetch both content/ and docs/ directories, and nothing else.

🧾 Conclusion

By leveraging Git's sparse-checkout (in non-cone mode), submodules, and Obsidian Git integration, we can build a streamlined and efficient knowledge management workflow. This setup not only reduces unnecessary clutter in local environments but also ensures version-controlled, automated, and scalable updates across platforms like GitHub and Vercel. Whether managing historical knowledge, creating new content, or organizing workspaces, this approach provides a solid foundation for maintaining a clean and productive knowledge ecosystem.

On the Burden of a Generation and the Pursuit of Self-Liberation

Mon, 21 Jul 2025 00:00:00 GMT

Today, I find myself standing at a crossroads — not just personally, but generationally. The world we inherited is no longer the world our parents thrived in. The economic engines that once propelled entire families into prosperity have slowed, sputtered, and in many cases, reversed. The age of exponential growth, globalization, and policy-driven uplift has passed. What remains is a system burdened by debt, over-leverage, and diminishing returns.

We — the so-called “bulls and horses” of this era — are expected to carry the weight of yesterday’s promises while being denied the means to fulfill our own. We are told to endure, to compete, to strive harder in a game where the rules have changed and the rewards have shrunk. But the truth is clear: we are not broken — the system is.

In such a climate, the traditional pursuit of asset-based wealth has lost its relevance. What matters now is the management of liabilities, the preservation of cash flow, and the cultivation of resilience. Every decision must be grounded in necessity, not vanity. Consumption must shift from the demand side to the supply side — from impulse to intention.

An iPhone 13 is sufficient. A secondhand car, if it meets one’s needs, is a rational choice. A NAS server, if it serves a clear purpose, is a valid investment. But anything beyond necessity must be questioned. We can no longer afford to be consumers of illusions.

This realization extends beyond economics. It is spiritual. Emotional. Existential.

Our generation is not only economically burdened — we are emotionally indebted. To our families, who meant well but often misled. To institutions that prioritized conformity over individuality. To relationships that promised connection but delivered confusion. I have carried these weights — guilt, resentment, disillusionment — but I carry them no longer.

I now see that awakening is not a moment — it is a process. A shedding of illusions. A reclaiming of agency.

The only viable investment left is the self. In language — to connect beyond borders. In technology — to remain relevant. In creativity — to generate value where none existed. In understanding — to navigate the chaos with clarity. These are the only assets that cannot be taxed, stolen, or devalued by policy.

I do not seek escape — I seek sovereignty.

I will not be a passive participant in a system that no longer serves me. Whether through strategic migration, skill acquisition, or intellectual reinvention, I will find a path to autonomy. Not to rebel, but to reclaim. Not to abandon, but to transcend.

I do not resent my origins. I am grateful for life, even if I critique the path it began on. I am thankful for the pain — it sharpened my awareness. I honor the love I once gave — it taught me what I value. But I will no longer be bound by inherited expectations or emotional debts.

The next decade must not be spent in regret or resentment. It must be spent in pursuit — of clarity, of competence, of freedom.

I am no longer asleep. I am no longer waiting.

This is the beginning of my true life — authored by choice, not chance.

By MatrixCore

July 21st, 2025