Laura Arjona Reina wrote:
> Dear l10n-english and i18n friends,
>
> Two of us (Spanish translators) noticed that some packages have their
> short description beginning with uppercase
> (https://packages.debian.org/jessie/magics++ ) and others with
> lowercase (e.g. https://packages.debian.org/jessie/libss2 ).
>
> 1.- Is there a canonical form? Which one?
The Debian Developer's Reference
("https://www.debian.org/doc/manuals/developers-reference/ch06.html#bpp-pkg-synopsis";)
says libss2 is right and magics++ is wrong.
> 2.- How to proceed with the packages that don't follow the rule? We
> would translate it following the rule, but how to report to the
> maintainer? Just file a bug?
It's only a very minor deviation from "best practice", so I wouldn't
file a bug report just for this. Mind you, the package description
for magics++ does make me wonder about a couple of other things:
# Executables for the magics++ library
#
# Magics++ is the latest generation of the ECMWF's Meteorological
# plotting software MAGICS. Although completely redesigned in C++, it
# is intended to be as backwards-compatible as possible with the
# Fortran interface. Besides its programming interfaces (Fortran and
# C), Magics++ offers MagML, a plot description language based on XML
# aimed at automatic web production.
#
# This package contains the MagcML binary that may be used with
# magics++.
(Note that there's some more informative text in the description for
libmagplus3.)
Some quibbles:
* it should expand "ECMWF" to "European Centre for Medium-Range
Weather Forecasts";
* "Meteorological" isn't entitled to a capital "M";
* "plot description" sounds like a job for movie reviewers, while
"automated web production" sounds like one for robot spiders - does
it mean "automated weather map production on the web"?
* the use of "that" to introduce the last clause implies that it's a
definitive relative clause, but I suspect that's wrong - it ought
to be a descriptive relative clause, introducing a binary *which*
can be used with magics++;
* is "MagcML binary" a typo for "MagcML"?
* what "binary" is it talking about? The package's file list says
it contains not one but three binaries (magjson, magmlx, and
mapgen_clip) - compare the plural "executables" promised by the
synopsis;
* what exactly would we be using this binary or binaries to do?
* why is the final reference to Magics++ uncapitalised?
These just about add up to something I might manage to submit a
wishlist bug report about, but I'd need to do some more research
before I could suggest a patch...
--
JBR with qualifications in linguistics, experience as a Debian
sysadmin, and probably no clue about this particular package
