All software is precarious
Over the last couple years, I replaced my Mac-only applications with alternatives that are available for both Mac and Linux. The first and biggest switch was using Firefox, and I’m really satisfied with it. I felt like Mozilla’s history in creating a browser for Linux meant I should just jump into whatever Mozilla was offering, including the bookmarking and read-later service Pocket.
I used Pocket steadily ever since. Admittedly, a read-later service is at least partially a place I put things that I’ll never read. I read around 20-40% while the remainder is pushed down the unread stack, becoming a museum of things I wish I had the intellectual curiosity to care about. Nevertheless, it’s valuable to have a priveleged group of web links, especially ones that are cleaned up and easy to read. It was a nice addition to my tools for consuming the web. And then this past July, Mozilla shut it down for good.
I’m not naive about corporate interests. I try to avoid apps and services backed by big tech companies. For any private, closed source, or for-profit services that I use, I’m already brewing an exit strategy. I’m familiar with the purposeful degradation of platform quality a.k.a. enshittification, and I’m comfortable switching services and platforms. But this hits differently because it’s Mozilla. In all the apps I’m looking to replace, Pocket was not under consideration, not even in the same universe. That’s why it’s frustrating. It feels like replacing Pocket is a project that fell on me out of nowhere.
Look, it’s not for me to say what a company should or shouldn’t do, especially when providing a free service. I know that these things have ongoing costs and require maintenance, but I assumed whatever motivated Mozilla to acquire Pocket in 2017 would lead them to eat much of its cost too. I’m left wondering, if I can’t depend on Mozilla for the long-term, who or what can I depend on?
I think this why self-hosting is having a moment, even if for a limited group of people. The drawbacks of self-hosting are obvious. It requires technical background knowledge, research, and maintenance. It takes time to do all of that, and when you do, you end up with something that is often not as polished or feature-rich as commercial alternatives. However, using feature-rich commercial stuff is no protection from rug pulls of one kind of another.
Right now I’m self-hosting wallabag is my solution. I haven’t used it long enough to form a strong opinion about it, but so far it seems like a perfectly good one-for-one replacement for Pocket. It has a browser extension, mobile apps for reading, and it was able to import all my exported Pocket data. I can’t ask for more than that.
But it’s no salve for my broader concerns about relying on software. There’s no escaping that all software is precarious. It’s not because it’s technically unreliable or unstable, but because the overlap between my needs and the provider’s needs is fleeting. It’s especially the case for commercial software, but it’s still true for community software or even independent one-person shops.
Over at 404 Media, Sam Cole was similarly frustrated when she learned Pocket was shutting down. A colleague suggested to her “copy-pasting links to articles into a giant document to read later”, and she concludes by writing “I might have to start doing that.” It’s not a bad idea. A simple file-based solution is more durable and easier to maintain than anything will ever be.
I’ve been thinking since Pocket’s shutdown announcement that if I had saved my read-later links in a text file, none of this would matter to me. A text file is immune to the changing expectations of software. As I plan my exit from other commercial apps and services, it’s a great time to consider whether they can be replaced with a text file, document, or even a spreadsheet.
Previously: