Re: [Monotone-devel] beginner questions

[Nathaniel Smith]

On Sun, May 30, 2004 at 02:35:50PM -0700, Adam Logghe wrote:
> Under the cool concepts dept- Ori Berger mentioned "friendly" hash names
> in his email.
> 
> http://lists.gnu.org/archive/html/monotone-devel/2003-09/msg00063.html
> 
> Its called mnemonic encoder-
> 
> http://tothink.com/mnemonic/

This has come up before... there's even a superior (IMNSHO) system
designed especially for Monotone :-), that I called "bibblebabble".
There's code for it in
  net.venge.monotone.gui/polytony/polytony/bibblebabble.py
and I have a design doc around somewhere I could put up if people are
interested.

I'm still a undecided about what should really be done with it; it
seems clearly more usable in a number of ways -- easier to read,
easier to remember, etc., and for solid cognitive reasons -- but
there's also implementation and cognitive overhead involved in
having two different representations for version codes; alternatively,
one could push this coding down all the way, so that hex ids aren't
even stored in the database.  But this would 1) bloat the database and
manifests (though the compression we do would help with that), and 2)
totally break every existing database, with the need to re-issue every
cert.  Though, hmm... switching to an "edge" or "context" based
ancestry system should do something like that too, so there might be a
window of opportunity there.  Hmm.  Maybe I should start lobbying
Graydon again :-)  I wonder how bad the overhead is, really...

-- Nathaniel

-- 
"But in Middle-earth, the distinct accusative case disappeared from
the speech of the Noldor (such things happen when you are busy
fighting Orcs, Balrogs, and Dragons)."

This email may be read aloud.