john/capsule
1
0
Fork 0
mirror of https://github.com/jbowdre/capsule.git synced 2024-11-22 13:12:19 +00:00
Code Issues Projects Releases Packages Wiki Activity

New entry via EchoFeed

Browse source
This commit is contained in:
jbowdre 2024-05-23 09:44:38 -05:00
parent 68c3c07833
commit 32fafdf26c
1 changed files with 25 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

25
markdown/incoming/a-comprehensive-evaluation-of-various-search-engines-i-ve-used.md Normal file
View file

@ -0,0 +1,25 @@
---
title: "A Comprehensive Evaluation of Various Search Engines I've Used"
published: "2024-05-23T14:42:41.000000Z"
updated: "2024-05-23T14:42:41.000000Z"
---
Call me old fashioned, but when I want to find out about something my first instinct is *not* to chat with a computer. I'm not interested in a conversational back-and-forth or a probably-at-least-partially-incorrect summary, I just want to find the authoritative resources that can tell me about a thing. So I use a search engine, and I've used a lot of different search engines over the past ~5 years or so.
Given that the world has just been recently reminded that so many of those ["alternative" search engines are literally just Bing](https://www.bleepingcomputer.com/news/microsoft/microsoft-outage-affects-bing-copilot-duckduckgo-and-chatgpt-internet-search/), I thought it might be time to throw together a complete list of the search engines which are out there. *(As with everything I write, this is a comprehensive and factual listing, and any omissions are deliberate and not because I'm lazy or not actually thorough in my evaluations.)*
- [**Google**](https://google.com). Used to be the default way to find stuff, then they started making the results [deliberately worse](https://www.wheresyoured.at/the-men-who-killed-google/), then they started shoving AI into every orifice. Hard pass.
- [**Bing**](https://bing.com). Used to be garbage, then started to get a little better (as Google results were getting a little worse), then they started shoving AI into every orifice. Hard pass.
- [**DuckDuckGo**](https://duckduckgo.com). It's Bing in a trenchcoat, but private. Meh.
- [**Ecosia**](https://www.ecosia.org/). It's Bing in a trenchcoat, but plants trees? Meh.
- [**Startpage**](https://www.startpage.com/). Bingle + Google (Bingle), but private. Meh.
- [**Brave**](https://search.brave.com/). Has its own crawler/index, and supplements the results with queries from Bingle - oh and the company's business model is [built on crypto](https://fossforce.com/2023/01/brave-a-great-browser-with-a-questionable-business-model/) and [related scammy behavior](https://www.pcmag.com/news/brave-browser-caught-redirecting-users-through-affiliate-links), and the [CEO is kind of a jerk](https://web.archive.org/web/20240504031305/https://www.nytimes.com/2020/12/22/business/brave-brendan-eich-covid-19.html). Hard pass.
- [**Kagi**](https://kagi.com). Own index, plus results from Bingle. Subscription-based business model, and honestly great results... but I've stopped using Kagi over [concerns about the CEO's character and the company's lack of focus](https://d-shoot.net/kagi.html). Meh.
- [**You.com**](https://you.com). Basically Bing + AI wrapped with a dumb chat interface. Pass.
- [**Perplexity**](https://www.perplexity.ai/). Basically Bing + AI wrapped with a slightly better chat interface, and nice inline references for the summaries. This is my go-to for when I want to ask questions for a topic, but isn't really useful for finding authoritative sources (documentation) directly.
- [**Stract**](https://stract.com/). The currently-most-interesting search project in the world. It's fully independent and [open source](https://github.com/StractOrg/stract), and offers a lot of control over how it works... but the results are kind of underwhelming. I want this to succeed, but I can't use it for regular search tasks.
- [**SearXNG**](https://docs.searxng.org/). A metasearch engine which pulls from every other search engine, and gives you knobs to toggle each source to fine-tune the results. It doesn't have its own index or crawler, but offers good-enough results without having to rely on any one (or two) provider(s). Also, I'm a sucker for [self-hosting](https://scribbles.jbowdre.lol/post/self-hosting-a-search-engine-iyjdlk6y). **This is my current default, at least until Stract gets better**.
If you think my evaluation wasn't comprehensive, thorough, or objective enough, you might be happier with this [more excellent comparison of search engines with their own indexes](https://seirdy.one/posts/2021/03/10/search-engines-with-own-indexes/).
=> https://scribbles.jbowdre.lol/post/a-comprehensive-evaluation-of-various-search-engines-i-ve-used 📡 Originally posted on Scribbles