Re: collatoral damage / regressions (was: page break failure)

[Hugh Myers]

Graham,

I wonder if it not so much a matter of prioritization as it is a
matter of intimidation. That is to say that most people seeing the
process from the out side find it a bit scary provoking a "oh I
couldn't do that kind of feeling."  I know that in my own experience
with users turned loose on the development problem, the single
greatest obstacle was overcoming their convictions of in-ability.
There are some exceptions--- most engineers seem convinced of there
ability to write legible prose, the problem is convincing them that
they can't ;-)

I do know that you've already discovered the value of enlisting your
user community, having said that, I think you might find them quite
willing to do more than you suspect.

As long as I'm exposing my thinking to the world (at least the
Lilypond world) another thing that needs to be pointed out to the UC
(User Community) is that they have a sure and certain value in their
ability to say that they don't understand a particular piece of
documentation. They may not know the answer to their questions, but
they do know the things that don't answer them. Such people, if they
can write and if they have "user empathy" make extremely good
documentation specialists, better than most programmers certainly.

I seem to be babbling again...

--hsm
p.s. to paraphrase Einstein, "Software should be just difficult enough
to use to be useful and no more so..."

On Wed, Dec 2, 2009 at 4:23 PM, Graham Percival
<address@hidden> wrote:
> On Wed, Dec 02, 2009 at 05:33:55AM -0700, Hugh Myers wrote:
>> Hadn't realized that a policy of
>> collateral damage was allowed. Given your approach to testing (and
>> your experience) this should work as well as more conservative
>> approaches.
>
> As Alexander wrote, it's not so much that collateral damage is
> "allowed", it's that we don't have the resources to ensure that it
> never occurs.  My fundamental job, above all else, is
> organization.  I match up willing volunteers (and sometimes
> pressure unwilling volunteers) with problems in lilypond
> development.
>
> There are few people willing to help track regressions; from that,
> I conclude that this is not a priority for users.  If anybody
> would like to help, please see:
> http://lilypond.org/doc/v2.13/Documentation/web/help-us
>
>
> I have more info specific to regressions ("collatoral damage"), if
> anybody would like to help with this.
>
> Cheers,
> - Graham
>