I see that Vassal 4 will be in C++, not GWT.
Will developing Flash games be as easy as doing Vassal modules? Anybody experienced in Flash game development? I do wanna put some games online.
Thus spake viewofheaven:
I see that Vassal 4 will be in C++, not GWT.
Only the core will be in C++. Scripting will be in Python.
–
J.
I thought scripting wasn’t decided yet
From: Joel Uckelman uckelman@nomic.net
To: messages@vassalengine.org
Sent: Mon, October 17, 2011 12:14:19 PM
Subject: Re: [messages] [Module Design] Flash game vs Vassal game
Thus spake viewofheaven:
I see that Vassal 4 will be in C++, not GWT.
Only the core will be in C++. Scripting will be in Python.
–
J.
Thus spake Timothy Mccarron:
I thought scripting wasn’t decided yet
No, but I’m leaning heavily toward Python over Lua at this point.
–
J.
Sorry, hijacking my own thread because it is now moot. I don’t think it is easy to code Flash and make it peer-to-peer playable. I want to make games that real people can play across vast geographical distances! That’s Vassal.
As for scripting language, depends what level we’re talking about.
The key idea is to “get away with as little effort as possible”.
At the highest level (end-users), we can use Lua. Or even code an easy UI with Python to babysit non-tech people doing scripting. Whatever the delivery method, end-users really don’t care, because they don’t do many complex things in their scripting anyway.
At the lower level, use Python (easier to code than C++)? For anything not as critical as graphics (cross-platform concerns), do it at this level. Python is certainly more robust than Lua for heavy work such as the VassalEngine.
At the lowest level, say for graphics, use C++. But which graphics toolkit? If truly lazy (and volunteers SHOULD be lazy!), maybe just go with Python (plus Tkinter). Or PyQt. And just pop down to C++ (Cython? Boost.Python?) when we hit snags (such as Java’s graphics snags we so abhor).
Has anyone thought of retrofitting parts of current Java-Vassal with C++ and a more competent graphics toolkit?
If I can throw in my silly 2-cents, I’d say go Python all the way first. We might be able to get Vassal 4 up within 3 months!
Key 2 questions to ask is: Can Python (or something similarly easy) do what Java can? Can Python (graphics cross-platform wise) do what Java can’t? Fix the problems we have first? Choose between Graphics Toolkits and C++/Python later?
Yeah, I know, long-term health of project and all. Last I remembered, I would spend 10 hours planning for something around a corner I haven’t explored, and 10 minutes to plan for something I see right before me. Iterative improvements please. We’re humans, not prophets! (I’m speaking for myself, I don’t calculate more than 0.5 steps ahead when playing chess. Yeah, true that human’s heuristics processing, or intuition, is something else altogether, so robots shouldn’t get smug.)
Also, putting Vassal up as an online browser-based service will mean immense browser R&D investments. In spirit of iterative improvements, let’s start small, and not browser-based? Browser may not be the right/adequate platform. Smart refrigerators now can tell us the recipe. But I’m not paying to R&D a refrigerator that cooks for me! Let’s just wait until GWT overtakes Python for what Python does best.
End this thread now? Put all scripting language discussion in proper thread?
Lastly, Vassal is great (and one-and-only, notwithstanding ZunTzu) as it is. Thanks to you all! The way forward will be to save real time for all you volunteers. That’s my vote. I don’t care if Vassal is using the most fashionable toolkits or not. I care that it works better and better with every iteration, without destroying the lives of volunteers.