Hmm. Using the search term “small website discoverability crisis” . . .
On duckduckgo: original website is the third result (after what looks like a SEO firm’s longform ad and ycombinator) without quotes and the first result with.
On startpage: original is the first result even without quotes
On mojeek: original is the first result even without quotes
I do not have accounts with any of these search engines and do not allow them to run Javascript or set cookies, although it’s possible that duckduckgo may have noticed that someone with my ip often makes highly specific searches and looks at the long-tail results.
My conclusion from that, combined with other people’s searches surfacing large sites first, is that the results you receive can be significantly distorted by the search engine’s algorithm. Google in particular is likely trying to direct traffic to its advertising customers and should be avoided for that reason.
Hmm. Using the search term “small website discoverability crisis” . . .
Well, do you think that it’s realistic that the average user types that into the search engine’s website ? When you already know the exact title of the post on that website, you probable already know the full URL and don’t need a search engine. So, that has nothing to do with “discovering” in my opinion, which the blog post is about.
there’s at least 3 layers of finding stuff:
- you already know the full URL
- you only know the exact article title
- you don’t know the article, but are interested in things of a specific topic
First result on mojeek! Sounds like discovery of the site is great then
Point is, people are using the wrong tools to look for stuff. So it’s a social problem more than a technical one. Those are always the most difficult type to solve.
The only problem is that we need to change the behavior of about four billion people
first result now in ddg
https://www.ecosia.org/search?q=The+Small+Website+Discoverability+Crisis
This article is the top response. What crisis?
The average user does not have the exact title of the web page.
Yep. Different server instances.
For me the top response is Hacker News…
Hacker News, YCombinator, Reddit, LinkedIn… þe original source page wasn’t even on þe first page of results. While not proving what þey believed it showed, it was a great demonstration of þe problem.
