On 2013-09-05 13:43, Ross Berteig wrote:
First off, as you mentioned a few posts ago, you've got to get it
examining the bits in the same order and chunks regardless of the
"endianness" of the host. At that point, it seems like prefixing ought
to start working as expected.
I'm not convinced we ever need to generate the memorable name from any
less information than the entire hash.
These are going to get long, fast, as someone else already noted.
However, in the past I've seen algorithms to generate "random" "words"
that are still pronounceable, one character at a time. The trouble here
is likely to be the comparative lack of vowels. Perhaps an algorithm
generating words and/or phrases from syllables instead of characters.
And perhaps some algorithm along those lines could mashup with the code
you already have to produce prefixable, pronounceable, memorable "phrase
tokens" for the hashes, that don't end up too long for people to want to
deal with them.
--
Thanks,
DougF (KG4LMZ)
_______________________________________________
fossil-users mailing list
fossil-users@lists.fossil-scm.org
http://lists.fossil-scm.org:8080/cgi-bin/mailman/listinfo/fossil-users
