Re: 2.18

[David Kastrup]

Keith OHara <address@hidden> writes:

> David Kastrup <dak <at> gnu.org> writes:
>
>> Ok, nobody including myself bothered reverting
>>     Break slurs between alternative endings in repeats; issue 1698.
>> in time for 2.17.16 and it is clearly unfit for releasing in 2.18 (new
>> functionality, new interfaces, change of behavior without significant
>> exposure, no documentation).
>
>> Two things I consider important for 2.17.17 are the strokeadjust changes
>> for PostScript output, and the vertical rest positioning.
>
> Is this parody?  April Fools Day was two weeks ago, David.
>
> Slurs across alternative repeats are a long-standing bug in old
> functionality (slurs, and repeats) for which users need not use the
> interface to receive the corrected output,

For some use cases.  For others it does not work, and can't work
conceptually.

> and the documentation change is to remove two sentences of
> known-issues.

When the change works as expected.  Which it does for some cases.

> Unfortunately, the patch did not fix the actual problem.

It depends on how one defines "the actual problem".  And that is exactly
the problem with making the patch part of a stable release.  It's code
not based on a coherent concept of how spanners are supposed to work in
the context of repeats.  There is no coherency in the code (instead, it
produces a lot of new spanners behaving ad-hoc), and there is no
documented manner in which the code is supposed to work.  It is not
verifiable, and it is not clear what behavior should be intended and
what behavior should be a bug.

And there has been no settling-down period where at least user problem
reports have started shaping the behavior (a poor substitute for
coherent definition and/or design, but at least a slight control
instance).

-- 
David Kastrup