It’s a bit unfortunate that my first blog post is essentially a rant, but there it is. I started to write this as a comment on Reddit, but soon I realized I didn’t want to give away free content to a platform that doesn’t value its users anymore (at least not the way it used to). But let’s not digress.
First of all: who am I? For the sake of this post, I think it’s sufficient to say that I’m a Linux user, as probably you are. After the infamous rite of passage known as distro-hopping, I finally settled on Fedora, which happens to ship GNOME as the default desktop experience—mind you, defaults matter. As a user, there are quite a few gripes that I have with GNOME’s default file manager Nautilus, and judging from the project’s issue tracker they seem to be shared by more than a bunch of people. I’ll list my top 5:
- Lack of type-ahead search1. Believe it or not, triggering a full-blown recursive search doesn’t replace the ability to focus a particular file in the current folder by simply typing (part of) its name. I suspect 90% of users—especially those coming from Windows or macOS—expect the latter behavior. As some folks like to say… if it ain’t broke, don’t fix it.
- Thumbnail generation is painfully slow2. There’s not much to be said about this. You can try it yourself by opening a folder full of pictures for the first time and observe how long it takes for all thumbnails to be displayed.
- Search is so slow it locks up the UI3. When typing a file name to search, sometimes the whole UI freezes for a few seconds. My limited experience suggests that some piece of synchronous code probably needs to be revised, but I’m just guessing.
- Icons keep moving downwards while thumbnails are being generated4. This is a good old one. Together with (2), I can only imagine how miserable photographers’ experience must be. Not to mention the lack of thumbnails in GTK’s file picker, which amounts to a meme now5. Since it’s technically not part of Nautilus, I wont’t dig deeper.
I understand GNOME lacks resources, but oh boy, when I see developers’ time and effort spent on endlessly redesigning and rewriting core apps, I wonder: is that really necessary? How about working on these major issues instead? One of them dates back to 2006. I guess that fixing a 16 years old bug is not as exciting as rewriting the default image viewer in Rust6.
What baffles me is that Nemo, Cinnamon’s default file manager, has none of the aforementioned problems. And guess what, it’s essentially what GNOME’s file manager used to be before the catastrophe that began with the release of Nautilus 3.6 sometime around 20127.
I hate when people shrug off the problem of Nautilus being an underperforming piece of software by simply saying: “It’s Linux, you’re free to use something else”. That’s not a solution, it’s diverting attention from the problem itself. I can’t wrap my head around the fact that such a fundamental and user-facing application of an OS is being almost neglected.
Maybe what I’m ranting about is an inherent consequence of community-maintained software, where a bunch of generous people contribute to what they want in their limited spare time. Users can complain how they want, but they can’t expect someone with the necessary skills and time to magically show up and fix such issues.
I don’t know much about the dynamics of GNOME as a community project, so probably this is a wild and naive take, but couldn’t the GNOME Foundation allocate some funds to a sort of bug bounty program to incentivize fixing major, long-standing issues?