Talk:Hash function

MD5 and SHA-1 weaknesses

This article should document the issues, raised last summer and sharpened last month, with MD5 and some other widely used hash functions. I'm putting a note that the algorithm has issues on this page and the MD5 page. If anyone wants to document this properly, look at:

• CRYPTO2004 Rump Session Presentations, was Re: A collision in MD5' by Jim Hughes, who is also cited in Tech Republic calling for MD5 to be phased out;
• SHA-1 broken and Cryptanalysis of MD5 and SHA: Time for a New Standardby Bruce Schneier;
• SHA hash functions already documents the issues with the SHA algorithms.

(I won't write this up properly: not my field and I have very limited interest in the topic) ---- Charles Stewart 15:20, 14 Mar 2005 (UTC)

I confused this page with Cryptographic hash function: sorry, no issue ---- Charles Stewart 15:26, 14 Mar 2005 (UTC)

Hash table searches

Relevant article in Quanta Magazine: https://www.quantamagazine.org/undergraduate-upends-a-40-year-old-data-science-conjecture-20250210/ Concerns refutation of a 40-yr old conjecture on worst case etc.Billymac00 (talk) 15:43, 11 February 2025 (UTC)

Dubious

The cited formula for a Chi-squared test seems to be plain wrong, see https://stats.stackexchange.com/a/532701/119748 MüllerMarcus (talk) 10:43, 22 March 2026 (UTC)

The formula was added https://en.wikipedia.org/w/index.php?title=Hash_function&diff=prev&oldid=917798238 (14:59, 25 September 2019) and appears to be wrong, as per SE. Replaced with standard chi-test. Darcourse (talk) 11:46, 9 July 2026 (UTC)