Re: [Monotone-devel] Re: user-friendly hash formats, redux

[Derek Scherger]

Nathan Myers wrote:
> Participants on this list are self-selected as tolerating the hex
> strings.  To any normal person, they simply scream "not ready for
> prime time".  A public announcement that we have done away with hex 

When I first read the monotone docs 6 or so months ago I think I had two 
different initial reactions. First, "wow, this sounds really 
interesting" and "hmm... those hex id's sound really ugly".

However, after working with it for a few days I had almost forgotten 
about the ugly hex id "problem". Since then I can't say that this issue 
has been a very significant one for me or that I find myself repeatedly 
groaning over. It does seem to be a psychological barrier to entry but 
it doesn't really seem to be much of a problem in practice, once your 
initial shock has worn off that is.

I beleive that the arch docs say something to the effect of "don't let 
our crazy names scare you off, they won't end up being much of a 
problem" and I think monotone could possibly take the same approach with 
hex ids.

Rather than trying to decode hex ids into pronouncable, but 
unrememberable things, how about when talking about revisions we talk 
about them in terms of their author and the date they were written? 
These aren't necessarily unique though and perhaps another cert on each 
revision containing a 1 line summary or name for the revision would help?

So rather than talking about 8511b72a1a65cd9ca269c1c5e661a90042a86143 we 
instead talk about njs' Nov 24 addition of the --full-revision option.

We might also want to shorten the display of dates to use the least 
amount of information necessary to be unique, which I think selectors 
already allow for. (something like the format that the unix at command 
allows).

just my $.02

Cheers,
Derek