Re: [RFC][PATCH v5 00/13][BZ 10871] Month names in genitive case

[Rafal Luzynski <digitalfreak at lingonborough dot com>]

31.12.2016 18:47 Zack Weinberg <zackw@panix.com> wrote:
>
>
> On Fri, Dec 30, 2016 at 2:59 PM, Rafal Luzynski
> <digitalfreak@lingonborough.com> wrote:
> > 30.12.2016 19:02 Zack Weinberg <zackw@panix.com> wrote:
> >> [...]
> >> FreeBSD using genitive for %B is a strong argument in favor, I think.
> >
> > That's true and please note that I don't mean just FreeBSD but
> > probably whole BSD family which also includes OS X, iOS and
> > probably more which I'm not aware of. Other reasons: [...]
>
> I fully expect that we will switch to genitive %B in a future release,
> if we take my suggested approach.

I think it would be pretty destructive to the users if we initially
implemented %OB as genitive and %B nominative and then swapped them.
Unless...

> I'm suggesting that 2.25.0 should go out with nominative %B and
> genitive %OB strictly as a matter of timing and consensus. As a
> mostly-uninvolved party it's my assessment that you have consensus to
> add %OB but not to make %B genitive. If %OB is in 2.25.0, we can swap
> nominative and genitive in a patch release (because that would be
> "just" a bug fix to the locale definitions), but if %OB doesn't make
> 2.25.0 you're going to have to wait a full release cycle (because it's
> still a new feature).

... unless you mean this patch release will be very soon after,
within few months, so that application developers and users
will not even notice this short period before the switch.
If that's true then your plan looks good.

My comment below may be relevant here:

> [...]
> > Indeed, those two solutions are not different regarding the source
> > code of glibc. The only differences are:
> >
> > - documentation: whether we say that ALTMON and %OB provide
> > standalone (nominative) forms and MON and %B provide full-date
> > (genitive) forms or the other way round;
> > - locale database: whether we put standalone (nominative) or
> > full-date (genitive) forms into the new alt_mon sections;
> > - applications: whether the developers will use ALTMON or MON
> > and %OB or %B when they want a nominative/genitive case.
>
> In 2.25.0, this would be documented as a new _experimental_ feature,
> and we would not make any promises about which %-code / langinfo
> constant corresponds to which case. In fact, we would specifically
> warn people that we might swap them in a future patch release, and we
> would ask for feedback from application developers.

That means that it doesn't matter if %B is genitive and %OB is
nominative or the other way round as long as there are no locale
data provided, no documentation, and no applications fully relying
that the results are correct and consistent.

Indeed, more feedback would be good.  Also I'd like to see some
feedback from native speakers of the languages affected.  I have
a feeling that I'm the only one here although sometimes I notice
some Slavic sounding names on this list.

> [...] We can document that one of the codes is for
> standalone month names and the other is for complete dates (which is
> which to be hammered out later),

Yes.  It can be mentioned in parentheses that often this means that
one form is nominative and another one is genitive.  But emphasizing
this nominative/genitive pair may be misleading because some languages
actually have nominative and genitive (actually, that's true even
for English) but don't use them in dates.

> and mention that using month names in
> complete sentences may require more information than the locale API
> provides.

To be honest, I don't find it necessary but maybe there should be
a kind of warning that although in English one can build any complex
sentence relying on glibc locale data it's often impossible in other
languages.

Regards,

Rafal