Owning Your Data

Juan Villela

This statement from Nelson’s recent post on Goodreads losing all his data hits on something that, although we all know to be accurate, we tend to fall short on preventing against:

Don’t trust any cloud service with the only copy of your data. Most companies are not quite so reckless but consider what you’d miss if an uncaring company lost your data.

And it’s true; most companies won’t lose your data this recklessly. That data is often their source of revenue, and they’ll do anything to protect it and reassure you that it is best placed in their hands. And therein lies the other problem; if your data isn’t at risk of being lost, it’s being utilized to violate your privacy.

There are many angles to come at this issue. And it’s something that’s been exhaustively talked about everywhere. So, instead, let me show you what I’ve been doing.

Roll Your Own

The solution to this is relatively easy to find. You need to host your data and publish it in the manner that best suits your needs. But, like anything on the web, there are many ways to do this. Let’s take one that should be approachable people who have published their blog at some point.

This site is built using Hugo. It’s one of the relatively newer static site generators. And like most of them, it can read data from common file types like JSON. That means I can maintain a data file that looks like this and use it to create my digital bookshelf of sorts:

[
  {
    "name": "Attack on Titan",
    "creator": "Hajime Isayama",
    "rating": 5,
    "comments": "What Game of Thrones wanted to be.",
    "category": "Anime"
  },
  {
    "name": "The Dark Forest",
    "creator": "Cixin Liu",
    "rating": 5,
    "comments": "Knowing it'll end, that it's coming, but you just don't know when. The anxiety eats away at you until you destroy yourself.",
    "category": "Books"
  },
  {
    "name": "Harry Potter and The Order of the Phoenix",
    "creator": "David Yates",
    "rating": 4,
    "comments": "It's your fault Sirius died, Harry.",
    "category": "Movies"
  }
]

From there, you can loop over each object in your templates and create a digital bookshelf to keep records of the media you consume, like, and review. And the beautiful thing is that you can make this look however you want. It can be clean and straightforward like Tom’s, or you can include cover art and filters like Dave’s to make it look more like an actual bookshelf.

But if the idea here is to maintain a massive shelf like Nelson’s 600+ books on Goodreads, then a single JSON file is not feasible. That’s where something like Airtable can come in handy. And you might be thinking to yourself, “Isn’t the whole point of this to get away from hosting your data on other companies’ servers?”. True. But the idea here is not to depend on third-party social platforms to share your likes and ideas. After all, unless you’re hosting your website on your basement’s home server, it’s likely hosted on some company’s servers. But can have better control over our data. And that’s the whole point: having control.

Gabriel had some excellent insight based on his recent journey with Obsidian:

One of the things I find fascinating (and refreshing) about the community of Obsidian users is how driven they are to build bespoke self-hosted tools. Even if Obsidian disappears or the plugins stop working, the data still exists on your own disk in files you control. Maybe I’m just old, but I’m tired of having to start from scratch because web services go away or pivot to something unrecognizable.

So, it’s that flexibility of making your data portable that we’re looking for. What I like about Airtable is that their free tier is quite generous (I have a mountain of data there and haven’t paid a dime yet), and their API documentation is quite accessible. But you could use Google Sheets if you need more clarification about Airtable. It’s an Excel-like tool that most people are familiar with and like to have access to. But also, their API is rich and allows you to pull any data.

Once your data is stored on one of these services, you can write a small script in the language of your choice to pull the bookshelf data and keep it locally as a JSON file that your website can read.

Why This Matters

Brad said it best:

Writing on your own website associates your thoughts and ideas with you as a person. Having a distinct website design helps strengthen that association.

Signing up for Twitter is free and easy. That is also true for Goodreads, Letterboxd, and any social platform. Although a lot of these platforms have done a lot to make your data a lot easier to export in recent years, the fact remains that you’re placing trust in a group of individuals who have little to no interest in you or the value you place on that data. I’m not saying we should all leave X for Y. But if you can take anything away from this post, it is that you shouldn’t only rely on one platform. Go ahead and review that book on Goodreads and fight with people on Twitter about your hot takes on it. But why not also publish it on your site?

See Also

Data Hoarding and the Death of Collecting

It's easy to lose sight of the physical relationships we once had with the media we consume in the digital age. The ease with which we can collect and store data has altered our relationship with media and how we organize it. We can reclaim this relationship by developing better systems for organizing and prioritizing information.

Spreadsheets Revolutions

I was dissatisfied with the current bookmarking solutions, so I made a custom one instead. This started with a myriad of Airtable bases, and then—for some stupid reason—I decided to roll my own hosted database.