Hubert Kowalewski writes:
> BTW, I'm gonna be a wet towel for everyone submitting new nations, but
> don't you think that the process of adding new nations is a bit out of
> control?

Whew, I unleashed something here. Jos has found this discussion, but it
ought to have been on -dev as well as -i18n, so that other nation
contributors have a chance to join in.

I think there's two things here: whether all these nations are enhancing
the game and belong in the core Freeciv distribution at all (ignoring
for a moment the impact on translators), and if so, how to help players
navigate them, and then there's managing the impact on the translation
team specifically.

Personally, I'm quite fond of our nation collection; I think it's pretty
cool that there exists this growing GPL'd set of nation descriptions,
with flags, cities, leaders, relationships (civil_war, conflicts_with)
and potted histories (from which I've learnt some stuff), of consistent
quality, all curated by a qualified historian.
I accept that's not necessarily a reason that it has to live in the core
of Freeciv, though.

I think this is a well-curated set, without duplication as some have
suggested; it's not a free-for-all where nationsets from random
contributors are incorporated without much thought. Much of the recent
work has been by one or two people, and you can see from discussion in
the patch tickets that care goes into the process. I think most
untidiness has to be blamed on the vagaries of human history :)

With regard to addition policy, there's a fairly detailed policy at
<>. I suspect that
Jos & co also apply additional criteria.

I'm definitely not for permanently throwing away the hard work of Jos,
Andrzej et al, by deleting nations outright, chopping down legends etc.
At most, IMO, some of the existing content gets moved into some kind of
expansion pack which there's no obligation to translate, but whose
continued existence is otherwise accommodated.

I agree that the current nation selection dialog is a bit overwhelming;
in the Gtk client, every nation group other than Oceanian and Imaginary
has a massive scrollbar even in full-screen. Some half-baked ideas to
make it more manageable:

 * More, finer-grained nation groups, and maybe subgroups ("Europe/Baltic").
   This would also help automatic nation selection; if I select some
   Baltic duchy, it'd be kind of cool if the game is specific enough to
   throw six other Baltic duchy AIs into the game with me.
   Requires re-classifying existing nations, and for period-based
   grouping, probably some scholarly discussion and compromise (see
   "Early Modern" discussion in <> et al).

 * More sophisticated UI for selecting nation groups. Being able to
   filter on the intersection of Ancient AND African, for example, to
   get down to a manageable list to browse. Don't really know how to
   make this intuitive (I'd try Ctrl-clicking on the nation group tabs
   to select more than one, but it's not very discoverable).

 * Make more use of the interrelationships embedded in the nations.
   I think there's probably some rich information in civilwar_nations,
   which the player doesn't get to see much of.
   (One thing I want to do but probably will never get round to is to
   put those relationships in Graphviz or something and see what the
   result looks like.)

 * Geographic selection: pick nations from a world map. Maybe augmented
   with a time period slider as well.
   (Maybe even start with the computer's current physical location, if
   there's a technically easy way of getting that.)
   Everyone wants to play their local nation, maybe they can learn
   something of the history of their area while they're at it :)
   Would require backfilling all existing nations with approximate
   coordinates and dates (where possible), and noticeable coding work.

If we are to be ruthless and somehow divide the nations into "core" and
"extended" ones, well, I have no idea where to start :) I'd want to
leave that up to Jos & co, if they are willing.
There are a couple of technical issues. One is that some civilwar/
conflicts relationships will likely cross the boundary (so that a
"core" nation refers to an "extended" one), and the game complains
(mildly) about references to non-existent nations, and we use it to
check for errors in nation files.
If the "extended" nations are a modpack, we want them to coexist nicely
with the "core" nations, so that when the extended nations are
installed, a "major" nation can still spin off a "minor" one by civil
war. That could be done by having the modpack replace the "core" nations
as well, or by the same team maintaining both sets in sync.
I'm sure these are quite soluble, possibly with minor code changes.

There's also the option I mentioned before, of keeping the nations where
they are, but reducing the sense of obligation translators have to
translate all the material. It would be easy enough to split out all
nation material from everything else and present two pot-files, and
document that translation of the nations.pot is strictly optional, to
encourage new locales.
(It would be possible to generate sans-nation stats even without doing
this, if that would help.)
But I can imagine translators not feeling their work is done if *no*
nation words are translated, because nation words pop up in the game.
Which would suggest we do need the "core"/"extended" split.

On some of the other topics that have come up:

re incompleteness/imbalance: I wouldn't judge by the current snapshot of
nations, since there's been a continuous flow of them for years. Jos can
probably comment further, but <> suggests that
they're approaching it continent-by-continent, so some areas are going
to be fatter than others while that's going on.

re in-game confusion caused by Poles vs Greater Poles etc:
conflicts_with already exists and is maintained to discourage nations
with confusingly similar names or flags from appearing in the same game.
I checked a couple of the examples listed as confusingly similar, and
they are marked as conflicting.

re links to Wikipedia: These will probably rot due to wiki notability
fads over the years, and need maintenance :(

re max players: IIRC, the people ran a game with the current
maximum number of players, 126.
(In general, I do think we need to look at some of the issues our
already expanded limits have brought to light before expanding further
-- although while city radius is a current issue and map size has been,
I'm not aware of number of players causing the code to blow a gasket.)

re "imaginary" nations: These are popular. I agree with Daniel; I don't
think we should get rid of them.

Freeciv-dev mailing list

Reply via email to