🚧 This prototype is actively under construction. 🚧   Please provide feedback at github/danielsgriffin/SearchRights/issues or via email to daniel.griffin@berkeley.edu.
  • 2023-09-08 15:08:02 Added to the ‘Is there a “searchable repository of examples” (Zamfirescu-Pereira et al., 2023)?’ remark for Perplexity AI:

    # 🔎 │ ask channel in the Discord channel: “ask \@PerplexityAI anything. Check pinned for usage.” (with a weekly challenge for the best question; pinned message show example queries (note these are queries of their Discord bot))

  • 2023-09-08 15:09:22 Added to the ‘Is there a “searchable repository of examples” (Zamfirescu-Pereira et al., 2023)?’ criteria comment:

    See also the LangChain Hub prompt repository (not searchable).

  • 2023-09-08 15:28:00 Added to the ‘Example searches’ criteria category comment:

    @mollick2023useful, not specifically re search systems, has called for “large-scale public libraries of prompts”

  • 2023-09-08 15:31:06 Added link to the ‘Is there a “searchable repository of examples” (Zamfirescu-Pereira et al., 2023)?’ criteria text: to “A searchable repository of examples…”

  • 2023-09-08 15:53:17 Added a paragraph re ragas under Scoping:

    I will not be trying to replicate metrics like that in ragas (tagline: “Evaluation framework for your Retrieval Augmented Generation (RAG) pipelines”; from Exploding Gradient). Those metrics are intended to examine: Faithfulness, Context Relevancy, Context Recall, Answer Relevancy, Aspect Critiques. I’m more interested in actions outside the RAG-pipeline itself. What can be done to improve coordination between searcher expectation and system performance? What can be done to remedy system failures at searching-time?

  • 2023-09-08 16:57:49 Added Criteria (incomplete)
  • 2023-09-11 16:12:08 Renamed Criteria (incomplete) to Additional Criteria under Consideration (incomplete)
  • 2023-09-13 09:15:02 Refactored text, table, and markings. Various smaller edits w/in.
  • 2023-09-13 12:06:17 Added example searches for Perplexity AI.
  • 2023-09-15 16:29:49 Added &tbm=youchat to the link for You.com in /criteria/can-you-search-by-url#you-com
  • 2023-09-21 14:03:18 Updated Phind’s mark for . I previously had it noted that you had to be signed in to share a search. I do not have documentation of that previous claim. Currently you do not need to be signed in.
  • 2023-09-27 11:53:34 Added a ‘Sharing searching’ criteria: Are share links (or links to searches) indexed and searchable in search engines (or the system itself)? (and made various UI changes).
  • 2023-10-05 10:26:14 Transferred from danielsgriffin/projects/csuirx to SearchRights.org and rebranded from CSUIRx to SearchRights
  • 2023-10-16 16:03:08 Added a new category: Supporting feedback
  • 2023-10-16 16:03:56 Added note re Perplexity AI’s API offerings in criteria/is-there-an-api-for-search
  • 2023-10-16 16:51:42 Added a new criteria in Standards & Openness: criteria/provide-open-source-models
  • 2023-10-19 01:21:36 Updated mark to ‘Yes’ for You.com in criteria/is-there-an-api-for-search; added to criteria details.

Works Cited

  1. Zamfirescu-Pereira, J. D., Wong, R. Y., Hartmann, B., & Yang, Q. (2023). Why Johnny Can’t Prompt: How Non-AI Experts Try (and Fail) to Design LLM Prompts. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. https://doi.org/10.1145/3544548.3581388