Content deleted Content added
→Rewrite: new section |
|||
Line 232:
==Infeasability==
The algorithm as given is infeasable: it does ''m'' mods (i.e. divides) including doing the exponentiation, to subtract the high character from the running hash, times ''n'' character positions in the text string = ''nm'' mods. Divide is microprogrammed on modern architectures, and is maybe 3-5 timers slower than multiply, and maybe 50-100 times slower than ADD/XOR/SHIFT. You'll give back all or most of the advantage over linear hashing. [[User:Sbalfour|Sbalfour]] ([[User talk:Sbalfour|talk]]) 23:13, 26 September 2019 (UTC)
== Rewrite ==
I've been a computer scientist and mathematician for 27 years. I can't work with this kind of topical and at the same time, minutely detailed presentation. The theoretical and technical presentation is absent. This article necessarily involves the GF(2) field and must discuss polynomial remaindering in the field. What's here is probably distilled from a high school lecture, and is the fabrication of the editors. It can't stand up to professional scrutiny. It doesn't give credit to those whose work is relied upon.
Proper mathematical notation and technical diction is entirely absent. We need to start over.[[User:Sbalfour|Sbalfour]] ([[User talk:Sbalfour|talk]]) 02:24, 27 September 2019 (UTC)
| |||