2016-08-23 3:42 GMT+02:00 Flaming Hakama by Elaine <ela...@flaminghakama.com>: > >> What hinders you *in trying* to create a minimum example? > > > 1) Because it is veers toward being a ridiculous and arbitrary criteria. > Speaking for myself, the example presented in this case was ***clearly*** > small enough to debug.
True, and my first step was reducing it to a minimal. > In my opinion, anyone saying otherwise either has a > tangential axe to grind, or is not familiar enough with the language to be > of help. Then I should unsuscribe from the user-list (...] > But to say things to the effect of "anyone helping you will have to start by > creating a minimal example" is uber-rubbish. disagreed [...] > If I were to critique the code in this case, I would rather emphasize the > awkward formatting: Line up your opening and closing braces, and use more > consistent indentation, and more of it. It is far easier to read pages of > well-formatted code than it is to comprehend a single convoluted expression. agreed [...] > 2) Because often much of the "non-minimal" code comes straight out of the > docs. What kind of culture suggests that quoting code based on the docs is > unfit for the basis of discussion on a user group? Code in the docs shows use-cases. On the user-list we discuss and try to solve problems. This is not the same. > Because minimal examples fetishize the minutia of lilypond while obscuring > the normal complexity present even in meager scores, such that it becomes > needlessly complicated to apply the fix. I find it counterproductive to > suggest that one should reduce an example to be smaller than what is > *musically* necessary--in particular, in terms of the number of staves or > voices in use. disagreed > Hammering on people to conform to the minimal example causes people to have > extra iterations on the list, and wastes everyone's bandwidth: > * a reasonable (or possibly non-reasonable) example, > * a follow-up (attempt at a) minimal example (it will never truly be > minimal), > * once a solution is suggested, a follow-up about how to solve the actual > problem in the first place, since it was not clear how to apply the solution > of the minimal example to the actual score, which has additional necessary > complexity. > > Wouldn't it have been better to just provide a response to the first > reasonable request? The user list serves (at least) two goals: - A user may get an urgent problem solved - follow-up readers may learn how to solve similiar problems This is done best with minimal-examples, for both cases. > 3) My question to people complaining about the non-minimal-ness of this > example is: Precisely which of these lines caused you to any extra time to > debug? See my first post in this thread. > This was not a rambling several screens of undistilled raw source, > but about two dozen lines, much of which is boilerplate. > > The things you took out to make it minimal: did you take these out just to > prove a point, or did you ***honestly*** think that removing, for example, > the names of staves and voices, key signature, clefs, or reducing a piano > staff to a parallel music expression would actually identify or solve the > problem with the duplicate time signature? Without reducing it to a minimal example I wouldn't have spotted the problem, i.e. \key creating a not grace-synchronisized Bottom-context. > I don't think that anyone in this discussion misunderstands *how* to create > a truly minimal example. It's just that there is an open question about how > relevant it is, especially beyond a certain point. > > Almost more importantly, there is also concern about the impact of how the > attitude conveyed on this list about the requirements for minimal examples > is a deterrent to cultivating the lilypond community. I should unsuscribe from the user-list. > I understand the intention of the requests (although not demands) for > minimal examples. But, as someone who has spent a lifetime developing and > debugging code, I can assure you that these demands are strictly > unnecessary, and come across as whiny and unprofessional. I should unsuscribe from the user-list. > It should be possible to encourage people to improve their code, provide > reasonable guidelines for submissions, and not come across as hostile or > insulting. If my replies came across as hostile or insulting, please accept my apologies. It was not intended at all. As a non-native speaker this happens sometimes to me, though. Cheers, Harm _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-user
