Writing ·
A directory is not an algorithm
Google killed the homepage. Not on purpose. Not because a meeting decided so. It killed the homepage the way a glacier kills a valley, slowly, with overwhelming mass.
The homepage used to be the point. You would go to a site, poke around, follow a link to a blog someone mentioned three posts back, and find yourself three hours later reading a stranger's thoughts about sourdough or sailing or post-war Polish cinema. Every click was chosen by a human. The chain was always a chain of hands.
Then search became the first click. Then feeds became the first click. Then recommendations became the first click. And somewhere in there, the person who used to find your blog stopped finding your blog, because the middle layer stopped routing them through it.
What a directory does
A directory is a list of things, kept by a person, grouped with intention.
It is the single most boring piece of software in the world. It is also an algorithm's opposite.
An algorithm optimises. A directory curates. An algorithm reorders a feed every day. A directory adds one entry in a week. An algorithm measures you against other readers and suggests. A directory ignores you entirely and keeps its taste.
That taste is the product.
Blogroll is a directory of personal blogs. A human reads every submission. If it is a real person writing real sentences on a domain they own, it goes on the list. If it is a content farm, a SaaS landing page, a paywalled Substack with four posts, or a tech agency's portfolio wearing a blog skin, it does not.
That is most of the work. The rest is grouping by where the writer lives, because place is an interesting lens, and lenses are what directories do.
This is not nostalgia
The usual pitch for directories is that they remind us of Yahoo in 1997. Warm hug. Brownies. dmoz.
That is not the pitch.
The pitch is that directories scale a thing algorithms don't. Trust. When a person reads every entry, you know what the list stands for. When an algorithm sorts every entry, you know what the algorithm benefits from, and it is usually engagement and rarely you.
There are places on the internet where algorithms are fine. Shopping, sometimes. Maps. Search for a specific named thing. There are places where an algorithm is a tax you pay for convenience. News feeds. Video recommendations. The autocomplete for your email.
And there are places where algorithms make the product worse. Places where the whole point is a person with a voice. Reading is one of those places.
What Blogroll is trying to do
Three things.
One. Keep a list of independent personal blogs big enough to be useful and small enough to have a spine.
Two. Sort that list geographically, so a reader in Lisbon can find five bloggers in Lisbon and email one of them. Real places are better entry points than "technology" or "lifestyle".
Three. Stay cheap enough to run and small enough to be run by people who read every listing. No venture capital. No ads. A ten-dollar yearly Feature slot if a blog wants one, which pays the hosting and buys the time to keep reading submissions.
The whole project is a bet that a few thousand real readers will prefer curated over ranked.
What you can do
Write on your own domain. See the other piece for why.
Submit your blog. It takes sixty seconds.
Read three blogs from a city you don't live in. Bookmark one. Email the writer. Tell them something honest about a piece of theirs you liked.
That last one is the whole point of the internet. The directory is a place to start.