If you want to create digital artifacts that last, they must be files you can control, in formats that are easy to retrieve and read. Use tools that give you this freedom.
Ā
ā Steph Ango, in Files over apps
This week I re-read Steph Angoās post about why we should prefer files over apps. The context over which it was reposted was because Notion bought Skiff, an email client, which was quickly followed by an announcement that the service is being sunset in 6 months (and its source code got moved to another repo with no warning; which made many people think it got pulled completely).
Thereās a ton of things to like about this idea of using open format files for everything. Itās pretty much the most guaranteed way to make sure your data belongs to you, because itās as close as you can get to physically owning something digital.
Nowadays, most of what we use has databases hosted somewhere we donāt own, and that data can only be read by that service. Some integrate with others, some allow you to export your data to other services (or even to files), but the truth is that you donāt have the same level of control. If that service runs out of money and goes down suddenly, or some stakeholder decides letting people export their data is bad for business, youāre at their mercy. Especially if itās a VC-funded company, which are encouraged to either go big or go broke.
Sure, thereās a reason those apps exist and we use them: dealing with files can be hard. It becomes our responsibility to manage them, keep them safe, and if you want to access them on multiple devices (your laptop and your phone, for example), then it gets a lot more complicated. So, is there a happy medium somewhere? I think so.
File-based apps
Steph Ango is one of the people behind Obsidian, a note-taking app that writes everything into Markdown (.md) files. It is a pretty powerful app, capable of linking and organizing notes in pretty complex ways. But still, all the data in there is stored in files that can be read by any program in any system. Even though Obsidian has come unique features (and even plugins that add even more features), you can use it with the peace of mind that, if anything ever happens to Obsidian, youāll still have 100% of your data with you.
Obsidian is far from the only app that works like this, which is great. It means that if you find another app that better suits your needs, you can easily take your data in there too. Off the top of my head, you could easily migrate to Logseq, Notable, or even VS Code without having to do anything.
And what conveniences do you lose with this kind of approach?
Well, syncing files between devices is not as straightforward (or at least not free). In Obsidianās case, you can pay for Obsidian Sync, which is actually what funds development. If you donāt want to pay, it can still be done in a lot of ways. Some people use iCloud or Dropbox folders, others use Git, or other solutions. But yeah, definitely not as easy, especially nowadays where we expect these things to be free. Which leads us to another topicā¦
Sustainable apps
Software costs money to build. A lot. Even if it doesnāt cost money to keep it up (i.e. there are no servers or support), thereās still all of the dev work that went into it.
What weāve been seeing happen a lot in the past decade (or even further back) is that startups receive huge investments, with the goal of growing as much as possible, as quickly as possible, to grab a good market share. So these startups operate at huge losses in order to gather a large user base. Then, eventually, the investors want their money back, and now startups need to figure out ways of generating revenue from all these users.
What happens then is⦠a lot of the things that brought users into those apps start being taken away. The things that caused financial losses to the startup arenāt really feasible anymore, and they need to either start charging for stuff that was free or hike up prices that were low. Or even worse, start making money off user data. The term āenshittificationā was coined the word of the year in 2023 for a reason.
This is because, as Iāve written before, VC-funded companies are encouraged to either go big or go broke. Thereās no middle ground. They canāt grow steadily at a sustainable rate because they got a big chunk of money that theyāve gotta pay back in a few years, and they gotta hurry. So, thereās even more reason to really consider not having all your important data on a server they own, with no easy way of backing it up somewhere else.
Which is why I have a big appreciation for companies that aim to build software (or anything, really) in a sustainable way. They start with products that are good and appealing, yes, but also with well defined monetization strategies. Companies like Plausible, the brazilian news blog Manual do UsuĆ”rio and the previously mentioned Obsidian come to mind. Iām sure there are more.
My next steps
I have a confession to make: I talked a lot about Obsidian here, but I donāt actually use it. I use Notion, a VC-funded note taking app that keeps all my notes on a database on their server somewhere. I also use some other VC-funded apps. So this post was written for myself, too. Often as I read some tech news I get a bit worried that I might lose some things. Eventually something will happen with Notion that will force me to change my workflow or make me migrate somewhere else (this migration, again, isnāt easy, because my notes are not just files).
So, Iām planning on changing some things. All my blog posts here are already just Markdown files, and Iām testing editing them in Obsidian for kicks (and maybe to eventually be able to post from a phone or tablet). This will be my first step into getting used to it and bringing over my important personal notes to Markdown files too.
Beyond the peace of mind of actually owning my stuff, itāll be nice to be able to admire everyone who builds the software I use, instead of being slightly concerned that theyāll turn on me someday.
Did this blog post change your life? Or maybe I made a mistake that ruined your day? You can always send me an email to tell me about it.