News for hackers. Stuff that matters.
Re:What software engineers should know about search (2017) by BiteCode_dev on Monday October 18, 2021 @02:57AM (#28902918)
Yes, but a good embeded search makes all the diff.

VueJS and Tailwind CSS both are indexed, but on those particular site I use the web site search _when I'm looking back for the reference of something I know_, because it's faster and more accurate than googling.

Granted, it's rare, but if you manage it, it's great.

Re:What software engineers should know about search (2017) by kristopolous on Monday October 18, 2021 @03:06AM (#28902961)
Maybe. The point in those examples was that they all solved search in a specific contextualized manner impressively well and despite that people are creatures of habit and they'll want to port the same generalized patterns over that they do for everything else.

Not consciously. They'll just do it and expect it to work which is why effective indexing and "SEO" (as in, the engine can scrape the content and crawl around without getting confused) is likely the actual work to do to implement "search".

Re:What software engineers should know about search (2017) by ricksunny on Monday October 18, 2021 @03:41AM (#28903125)
From my experience yesterday I can tell you that Zenodo is struggling with reliably repeatably returning results basic boolean search queries. I think I’ll try parent’s search-the-site-via-Google approach.
Re:What software engineers should know about search (2017) by svcrunch on Monday October 18, 2021 @11:57AM (#28906882)
Agree with your sentiment.

I cofounded ZIR AI to provide ML (vector)-based search as a PaaS solution, similar to what Algolia or Elasticsearch do for keyword searching.

We have a demo (https://zir-ai.com/demo) running over Quanta Magazine articles. Not only can it outperform the keyword search embedded on quantamagazine.org, but one some queries, it even outperforms Google with a site restrict (e.g. how old is the universe site:quantamagazine.org).

> Granted, it's rare, but if you manage it, it's great.

ML-powered search will make this commonplace, once the tech goes mainstream in the next 2-3 years.