I quit search engines for three months.

Yup. Three months without search engines. It was… not without issues.

If you were following me on my Mastodon (until I started taking a break from microblogging), you may remember that I quit using the crawler-type search engines (like Google, DuckDuckGo, Startpage, Lycos, etc) at the beginning of the year… sort of an accidental New Year’s resolution, but more of an experiment in how my brain works.

(No, you pedants, this does not mean I didn’t “search” using the built-in search features of specific websites like Wikipedia/forums/blogs/etc… just keep reading to learn more.)

Well, today I finally gave up and searched something. The experiment may be over, but I learned a few things.

80% of what I’m looking for can be found in Wikipedia or on forums.

That’s no joke- I would guess a good 80% of my search terms pre-experiment were just me landing on Wikipedia, Reddit, or the forum for whatever project I was working on.

During the length of my experiment, I used Wikipedia as the default search engine for my browsers (mainly to catch habitual searches). It probably handled a third of my searches just fine, without incident.

As a (now former) sysadmin, my next go-to has been Arch Wiki– probably the best example of Linux documentation in the world. Even as a Debian user, Arch Wiki really can’t be beat.

Beyond that, I started to get into forum posts again. Bookmarking the various forums and jumping into message lists was vital. A lot of the projects I use everyday have robust forums with lots of answers. I don’t think I could have finished my Chrultrabook escapades without forum.chrultrabook.com– a really shining example of an open source project with a great forum.

And then, of course, there’s the social-media-esque sites like Reddit and Discord. Unfortunately a lot of projects use these tools, and I found myself making much more use of them than I would like. (I’m not going to shame any of you here, but if Discord is your support system, please get a searchable forum.)

I was able to replace probably 80% of my searching just by bookmarking a whole lot of forums and relying on well-maintained wikis.

Dr. Manpage or: How I learned to stop worrying and love the docs.

That last 20% though- that’s really where things get dicey.

The project docs for my various tools are incredibly important to me now, and in various states of order or chaos.

Referencing documentation frequently has been a welcome change overall. Man pages (manual pages built into Linux/UNIX/etc systems) are vital to how I get work done now. Instead of hitting up a search engine to find the answer about whatever system I’m working on, my instinct is now going to the terminal.

Of course, that doesn’t work for everything. But most projects do have some level of documentation, and I’d cite as a good example the Open Source Cartridge Reader project that I mentioned a few weeks ago.

Not every project is so lucky. Some projects have very poor documentation. And this experiment has highlighted (for me at least) the need to focus and fund documentation for the projects we rely on. I don’t want to name names here, but in general- please support those who write and edit the docs.

Actually talking to people.

Instead of DuckDuckGo helping me get right to it, I had to [gasp] pay attention to where folks were getting work done for the projects I use everyday.

This has been a net positive for my brain space, as I’m a pretty extroverted person and I like chatting with folks. Instead of searching for an answer right away, an answer which might be laden with inaccuracies and SEO nonsense, I was getting a chance to ask folks how to solve problems. Forums, group chats, Mastodon, that sort of thing.

Forcing myself away from the answer oracle and toward actual humans who have opinions? That was worth the experience.

Slowing down: definite downsides.

So, all of these positives, surely I think everyone should quit using search? Absolutely not.

For one, while I was able to mostly get along finding information I needed, it was far from foolproof. Plenty of wild goose chases have ensued. I have spent the better part of two days triaging a problem with my jog shuttle in Kdenlive- turned out I forgot to set permissions, as outlined by this helpful blog post I found in my first search today.

I have no doubt lost momentum in my professional and personal life because every time I have to find information online, it takes longer than it used to. It’s taken me longer to produce videos and writing, it’s taken me longer to find where to go for Thai food, it’s taken me longer to discover new tools.

This experiment may have cost me opportunities, but I’m still glad I did it.

A more personal web.

When I finally figured out what was wrong with the jog shuttle, instead of just going about my day, I reached out to see if the Kdenlive docs could use a blip about it.

I wouldn’t have done that before this experiment.

Lately I’ve been reading about digital amnesia, something I’d like to explore more in a future post (or perhaps a video). I’m a sample of one, but I feel like this experiment has helped me get closer to the sites I use everyday. Instead of a generic “answer machine”, I have a collection of links that I have to curate. Resources I need to parse in my brain. Stuff I have to remember.

Breaking 20-year-old search habits has been useful to me, and I think I’ll continue doing periodic “search fasts” to identify other areas where I have useful first-line tools for my brain.

And I’ll probably leave Wikipedia as my default search engine. At least for a while longer.