The ongoing Reddit blackout in protest of Reddit’s API pricing has an unlikely or unexpected victim, that being Google. This is because, similar to Wikipedia, Google drives a lot of traffic to Reddit from people using Google indirectly as a Reddit search engine.
The quality of Google’s search results have gone downhill precipitously over the years, as many people can attest:
Indeed, a Google search about the very badness of Google search, yields over 19 million results. To improve the quality of results, people have devised various hacks, most famously, appending “Reddit” or “Wikipedia” to search queries, or using lesser-known search engines such as Brave:
now that appending searches with "reddit" no longer works for targeted queries, google is officially useless. reddit blackout has highlighted the fact that google is just a propaganda and data farming site.
what is everyone using for alternative
— Schwab (@realhumanschwab) June 13, 2023
My own experience is that Google works best for obscure queries, especially academic papers and stuff from years ago, such as old math papers. The more obscure the better. When writing my papers, if I needed to know if something had been done before, I would run it through Google instead of arXiv, because I know that Google indexes all arXiv results anyway, plus almost every other journal. But for more mainstream stuff, using Google is like rummaging through boxes at a garage sale and trying to find something that isn’t broken.
There are two key problems with Google search, which a competitor like Reddit, Microsoft, or Open AI can capitalize on by fixing, those being stale results and vague / irrelevant results.
‘Stale results’ are results from months, even years ago, showing up as “recent” due to some Google bug, or sites exploiting Google’s update/relevancy feature. Sites that have fresh, updated content tend to rank higher than sites with old content. Sounds good in theory, such as for topical matters like news, but webmasters soon figured out they can keep perpetually ‘updating’ old stories to have them rank higher, such as stories from as far back as a decade ago showing up as recently as last week. An ‘update’ can simply mean updating the “article:modified_time” to the present date, without having to change the actual content of the article, as shown below:
Vagueness is harder to pin down. Consider a more mainstream query ‘does volume eating work’ (no quotes), as shown below:
Notice how the results explain how it works, not if it works. The word “work” is not even included. “How it works” vs “if it works” are lexicographically similar yet semantically distinct. Many people know what volume eating is, but they want to know from actual people who have tried it, if it’s effective for weight loss. And that is where Reddit is useful. Reddit comments are written by actual humans who are either looking for answers, or are supplying incisive answers or explanations, not filler. And most importantly, based on unbiased, first-hand experiences.
Some people argue that blogs are the answer to a more centralized or impersonal web. I disagree. Reddit answers and comments are better than blogs, which are not uncommonly full of ads, SEO-spam, signup forms, and generic and vague copywritten filler. Especially personal coaching blogs and business blogs, which almost universally have generic, bad content, and fill up Google search results for anything having to do with health or finance/money. Not to mention, many bloggers are paid to promote products and are not fully honest, or do not use the products they endorse, compared to the Reddit users who tend to be much more discriminating. If something does not work as promised or has some sort of ‘gotcha’, you will hear about it on Reddit first, not a blog.
If you want to know if interment fasting works, instead of reading a vague blog about the ‘benefits of fasting’ or ‘what is fasting’ (you know what the benefits are, or else you would not be searching for it), or having to sign up to a newsletter, why not read people’s actual experiences with it. This is what Quora, in theory, was supposed to do, until it too succumbed the general ‘badness’ and spam that has afflicted much of the web.
Rather than trying to squeeze developers with unreasonable API pricing and driving its users to boycott in protest, the opportunity is open right now for Reddit to form some sort of major partnership to monetize or license its huge trove of user-generated content, or at least take market share away from Google (which makes Reddit valuable in its own right even if the content is not monetized). The market demand cannot be more obvious than millions of people literally appending your brand when doing search queries on a competing website. You don’t need focus groups or surveys to see this.
Maybe instead of Reddit.com , users are redirected to a landing page that incorporates Reddit user-contributed content, but it’s not Reddit. The landing page, using AI, rewrites and organizes the answers and threads in a more readable format, such as by correcting spelling and grammar.