As someone with one foot in the UMC world and one foot in this world,
I think the most logical path is a break between 1.0 campaigns and
1.1->2.0 campaigns. I may be wrong, but I am operating under the
assumption that the 1.1 vector towards 2.0 is going to result in a
potentially vastly different game than 1.0 is. I would encourage
players to understand this difference and to choose which WML
featureset they will use.
It sets a reasonable expectation on potential campaign writers and
frees the development team from any restrictions. For example, I
think there's a plan to shuffle file locations, and I won't believe
any assurance made now that it's over. If I make an [item], do I use
the 1.0 image location or the 1.7 image location? What if the image
or unit (sea hag) is present in 1.0 and removed/changed in 1.3 (like
Jetryl's new story images)? Maybe I include both tags? Maybe I
duplicate every resource I use just in case it's changed? How do I
balance such a campaign if the stats are differenet? In cases like
these it may not be possible to write a campaign that is restricted
to 1.0 features but still works in 1.x -- you just have to choose. I
don't think there's anything wrong with this. I wouldn't expect a
Warcraft I scenario to work in Warcraft II either. For example, I
plan to make Liberty a 1.0.x-only campaign and create a separate 1.1+
version only if there is demand for it.
Turin's idea for multiple servers addresses the fact that 1.0 clients
are not necessarily going to have an updated campaign server client
that allows filtering. If content and version filtering are added to
1.0.3 (wishful thinking!), you have tons of players with 1.0.0-1.0.2
with no reason to upgrade who are going to start seeing incompatible
campaigns. It's frustrating enough to have broken campaigns, and now
you are adding to the problem. You can require that publishers of a
1.0.x branch use the correct 1.0.x version in the server code, but it
would be nice to keep content filtering in 1.0.x an optional update
rather than an essential one. It fits with the 1.0.x philosophy as I
understand it (where the value of x doesn't matter - much).
Turin's idea for multiple servers also requires the least amount of
work, I think, but I could be wrong. If you leave the 1.0 server
available for 1.0 clients, you free yourself to make changes to
whatever without having to worry about the impacts on the 1.0 clients
that we hope people will be using for a long time.
Scott
- Re: AW: [Wesnoth-dev] Two campaign servers? Scott Klempner
-
