From senator-bedfellow.mit.edu!bloom-beacon.mit.edu!nycmny1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail Thu Mar 21 11:22:24 2002 Path: senator-bedfellow.mit.edu!bloom-beacon.mit.edu!nycmny1-snh1.gtei.net!cpk-news-hub1.bbnplanet.com!news.gtei.net!feed2.news.rcn.net!feed1.news.rcn.net!rcn!not-for-mail From: "Brian Payne" Newsgroups: rec.arts.int-fiction Subject: Re: [Inform] Map creation utility Date: Mon, 10 Dec 2001 09:22:34 -0800 Organization: Organization, me? Hah! Lines: 151 Message-ID: <1008004854.154232@sidehack.sat.gweep.net> References: <1007765094.759308@sidehack.sat.gweep.net> <98ef019f.0112100509.6615125@posting.google.com> X-Trace: UmFuZG9tSVardDAAsiL2pevLRrzrDA4jDI87KDap//J0WQtqfmUdJ8LWpWVyqIxv X-Complaints-To: abuse@rcn.com NNTP-Posting-Date: 10 Dec 2001 17:20:54 GMT X-MimeOLE: Produced By Microsoft MimeOLE V5.00.3018.1300 X-Priority: 3 X-MSMail-Priority: Normal X-Cache-In: nntpcache 2.4.0b5 (see http://www.nntpcache.org/) X-Newsreader: Microsoft Outlook Express 5.00.3018.1300 Cache-Post-Path: sidehack.sat.gweep.net!unknown@206.63.156.65 Xref: senator-bedfellow.mit.edu rec.arts.int-fiction:87810 "Plugh!" wrote in message news:98ef019f.0112100509.6615125@posting.google.com... > As a professional sowftware developer, I can assure you that the #1 > most important aspect of any software development project is > specifying it. Believe me, you will make your life a *lot* easier if > you first decide exacly what you want to implement, rather than > beginning coding and deciding 1/2 way through that you'd like to > handle the rooms contents too (item, actors & scenery). As a professional software developer myself, I completely and utterly agree with you. :) I have been down the unspecified road far too many times to want to set foot on it again, especially for something that will likely draw more fire than praise. :) > > So, do yourself a big favo(u)r and ask your potential users what sort > of a tool they would like to see. My biggest nightmare is that I go > away & hack code for a year, announce the result to r.a.i.f and > discover that everyone wants it to work differently. Canvas users up > front, write a design specification, then (finally) start coding. > Well, now that you've mentioned it, what would YOU like to see? I think what I'll do is post the general design that I've come up with so far, and let folks tear into it. And, now that I think about it, I should probably ask for name suggestions, since I'm drawing a blank on that end. > Are you going to have a fixed grid? Will North always go up the screen > & south downwards? How would that handle Colossal Cave, where you can > go North from A to B and return by going North? Fixed grid BAD. Ugh, no. As for North always being up... well, yes. In the case of Colossal Cave, though, you'd see something like this (pardon my horrible ASCII art): +--------------------+ | | | | A B or (more complex): +-----+ +-----+ | \ / | A x B <-- lines do not join at the x | / \ | +-----+ +-----+ You see? In either case, North from A takes you to B, and North from B takes you to A. The only difference is the, ah, entry vector. :) Which, unless there's a door involved, doesn't matter, really. > > Someone else in this thread rightly mentioned directions like in/out. > What about ssw, etc? or magic words which move you from one room to > the next? For the first beta, at any rate, stuff like that would be undefined by the GUI mapper. Simply put, this is intended as a tool to help you get to the enhancement stage faster -- ie, to let you breeze through the generalized map layout and room creation, and THEN go back (by hand) and add in the niftiness. > > I decided that I would have to offer totally flexibility otherwise > there would be bound to be some unsatisfied customers. D/L my current > work in progress & see how I handled it. But you *must* have this sort > of thing decided *before* you write your first line of code. > Otherwsise, it will all end in tears.A good idea would be to post the > design here for critique & feedback. Good call. I'm still working on the initial spec, and testing my concept of graphics routines (if I'm going to make this portable, I can't rely on, say, Windows event calls to determine when the user has moved the pointer over a room, for example), but I'll post a 'features list' for everyone to review in the near future -- later today, maybe (depends on my work schedule). > > If you decide to do more than just the map, there was a promising > project called Visual Inform, which I think may no longer be in > development - see http://www.iflibrary.org/VInform/ - perhaps you > could take over the reins of that? I looked at it, but it's a bit beyond the scope of something that I'd have time for as a hobby project. Besides which, I always find it harder to take over someone else's project (even assuming that they'd let me) than to build my own. If it turns out that I do have time later on, though, and Visual Inform is still under development, I may offer my services as a UI designer. That's what I specialise in during my day job, and what I enjoy the most. > > Now, as to programming langauge ... you want something with gui GUI > capabilities and preferably something cross platform (although a > windoze solution, used with Wine would help Linux users, but not Mac > users). Java sounds the obvious solution, if you don't think that it > would run too slowly. I believe that Borland do a good free Java IDE. I'd be better able to debate this is if I knew more about Java; however, from my experience (admittedly, not much), Java wouldn't be too slow, but it'd incur a much longer development cycle. However, I *will* be releasing the source once I've gotten something put together to release ;), and if anyone wants to port it to Java, they're more than welcome to -- I'll even help, as much as I can. :) > > Personally, I use Borland's C++ Builder. If you don't like C++, but > feel that you could handle Pascal, then Borland's Delphi for Windows > would be perfect and they give away a free version, named Kylix, for > Linux. Code written for one should pretty much run unchanged on the > other (I'm awaiting the release of the Linux C++ IDE). I like Borland (always have, since Turbo C 2.0 ;), I use Microsoft at work, and since I'm just a poor broke lil programmer, I use DJGPP at home. :) I've never used Delphi, but it sounds like I may have to check it out. > > Although my project is for TADS, I would be more than happy to point > out potential pitfalls, share design tips, etc > > But, first, ask your self if you really have the time & energy to > devote to such a thankless project. Quick answer? No, of course not. Who does? But I'm still going to. I have *fun* doing this sort of thing -- you could, assuming you knew her, ask my fiancee about it: I've been bouncing off the walls at home the past week while ironing out the most glaring of the problems I've defined for myself. Heh. This is *fun*... and fun is something that I can never get enough of. Brian Payne sofaspud at sofaspud dot org