Talk:Source
From OHRRPGCE-Wiki
Mike: I forecast Rusalka to be released on April 10, 2006, based on the current pattern in the table above.
Bob: I just changed the release goals for Rusalka. There is enough cool new stuff already implemented, that it is time for a new stablee release. Of course, I won't refuse for future goals, it just means I won't wait for them.
TMC: Why estimate when you can extrapolate? Take a look at this old graph, and realise that the last update was May 19, so that last bar would be ~330 days long on an updated version.
There was obviously a change after 2001, so we'll ignore everything before then. Clearly if we want to skew our results we should exclude wolfwood as well. :D Lets say it's increasing exponentially. Afterall, we don't want to finish too early before our 2137 deadline. According to the exponential regression function on my calculator, we should expect the next update 316 days after the last, or about March 31st 2006.
As for quaternion, don't forget the massive amount of documentation missing. That'll keep everyone busy while I make final suggestions. Or did we write it already?
Bob: *laugh* I think that chart would look better plotted as a timeline rather than a bar-graph :)
Mike: Yes, not to mention that the X axis is backwards. But... My forcast was based only on the release dates of Paternoster and Quaternion. Besides, isn't it a known fact that previous behaviour does not indicate future behaviour?
Heck, we're pretty close to releasing Rusalka. But, alas. the bane of the programmer (especially open-sourcerers): documentation... "I know how it works, isn't that enough?"
Bob: As far as new plotscripting command documentation goes, I think we are actually in really good shape already. Most (maybe even all) of the new commands already have entries in plotdict.xml -- as for new features, well, we are no worse than the usual past practice of not documenting anything at all outside the whatsnew.txt file ;)
Mike: Heh, looking back at the previous discussion, I realized that my original estimate was only 11 days off. :P
Anyway, I must protest the next code name 'simonsays'. That's just stupid. I agree with the concept behind it (Simon says 32-bit!), but that's a poor execution. We need to leave our 16-bit past behind, and embrace all 32-bits of the future! Maybe even 64, but there's no need to rush things.
The codenames for the OHR have usually been long words that don't mean anything (well, except Handshake, but that doesn't count). We should continue with this. I can't think of anything right now (my brain doesn't function too well at 3:00 am), but I'll try.
Bob: heh, well, it was just an impulse. I am open to suggestions.
Mike: I hereby nominate Jobbernowl, an archaic insult.
Bob: That is an awesome word! ... but since ozarks, names have been alphabetical.
Mike: ... you're right... I shall continue my search then.
(Five minutes later)
Oooh, found one. It makes sense, and starts with "S". I think you'll agree that it would be an awesome name when you hear it: Serendipity
Bob: hehe. That reminds me of this
raekuul: Dosen't O come before P? didn't Paternoster come before Ozarks? (Or am I wrong about the releases?)
Mike: It's funny. Every version we make, the same three features (plotscripting arrays, customizable menus and precise weapon placement) keep moving down to the next version. Why don't we just throw in the towel, and say that they'll happen when they happen?
Bob: yeah, they will happen when they happen. They are just on the goals list because they are clearly defined objectives that we have already committed to implement... eventually.
raekuul:...Then "Upsilon", then "Virtuoso", then "Wimbly", then "eXtremely (modified)", then "YetAgain", then "Zoogeophysical"... naaah...
Arpgme: Hasta-la-qb??? Didn`t you drop the Qb source for Fb? Also, will I be able to hack the ohrrpgce source if I have experence with Python?I plan to clean that big ball of mud.I want to put it into python and make it run in windows.
The Mad Cacti: Actually, we hacked up the source real good so that we can compile with both QB AND FB, so we could continue to make DOS versions (that's why there's two downloads). This was an important thing to do, so that people could continue to use the old build when they ran into problems with the FB one (and they did). However, it also means that we're nearly as limited as ever, because we can only make additions which can be implemented in both QB and FB versions (exception being midi music). Now we're going to stop supporting compilation of the source with QB, opening up wonderous possibilities.
What do you mean by "making it run in Windows"? It already does, prehaps you're thinking of the fact that it runs fullscreen? The FB version can run in a window, it just doesn't by default. However, have a look at Jormungand, the effort to rewrite the OHR in Python!
Bob the Hamster: notice the new development plans section I added. Mostly this is a place for us to either (A) organize thoughts we have already discussed on the mailing list or forums or other places, or (B) to lay out personal projects so other developers get a chance to comment (for example, I haven't talked much about my Plan for floating rectangle interface but anyone thinking about working on related features like ui customization ought to be aware of it now)
