Date: Fri, 24 Feb 1995 04:15:40 +0900 From: Stephen Turnbull To: huntercr AT cs DOT purdue DOT edu Cc: DJGPP AT sun DOT soe DOT clarkson DOT edu Subject: RSX is great and the point of freeware is questioned... Everyone seems to be missing my point a bit... and if the issue brought up cannot be resolved in another way, I think RSX's use becomes limited. I don't think we're missing the point. We don't like the situation any more than you do (well, actually, I don't really care as I have never written or ported a Windows program, and don't intend to start). Everyone is saying that the requirements for using RSX for djgpp are to have a commercial compiler that is intended for windows... Yes. That is about as silly as saying that if you want to use djgpp for 32-bit dos apps, you need to buy Watcom! IMHO one of the great Tell it to Bill. He will laugh all the way to the bank. points of djgpp is that it provides a 32bit compiler for those of us who can't afford a commercial one, and if RSX requires files from a commercial product, then it's use becomes restricted. Yes. Tell it to Bill. He will laugh all the way back from the bank. I am sorry if I sound like I am "ungrateful", but it seems to me that if windows.h can be found in ~any~ commercial compiler, that it would not be too difficult to get it included in RSX. *You* can certainly steal it. But it *is* stealing. I think you *should* sound "ungrateful"; just direct it in the right direction. At that guy who seems to spend an awful lot of time in the bank. (It is not clear to me that Bill Gates has done anything evil. However, you don't owe a $billionaire "gratitude"; he's shown he prefers hard cash to thanks. Instead, keep buying and producing Windows products, and keep the royalties flowing. Save your gratitude for DJ et al.) After all.. what could possibly be included in that header? If it is "(C) 19xx, Microsoft Corporation." only prototypes, defines, etc, then since there are no external libraries from these compilers cannot one assume that the coorresponding functions writtenfor these prototypes are *not* taken from the commercial product? If this is true then the header should be able to be included since the file does not point to the copyrighted source anymore. What do you think headers are? *Headers are source files*. It is only by programming custom that they don't contain executable code. But even so they often contain macros, and in C++ they contain inlined member function definitions. That's beside the point: even the prototypes should be copyrightable. I would be willing to bet that if you look it up, you will find that the Windows API itself is in fact copyright Microsoft. And you will find that all of those commercial compilers have licensed the use of that API. (Bill would be breaking an implicit agreement with them if he gave away the API to RSX, you know.) Read the COPYING file that comes with DJGPP (not COPYING.DJ, which is relatively straightforward, I mean the stuff that RMS puts out). I suspect that you will find that it may be *impossible* to simultaneously satisfy the COPYLEFT and Microsoft's copyright in a single distribution. Now, even if that copyright is bogus, do you want to try fight another look and feel battle? And in this case I think that Microsoft is on much firmer ground than the plaintiffs in the look and feel cases. Good luck to you. I'm afraid the Devil or Daniel Webster would be more help tho'. If you don't like this state of affairs, you can go Linux or *BSD + XFree86 (all truly free software), or DJGPP + DESQview/X (which costs $150 or so, but you can get the DJGPP libraries for free, Motif cheap; XFree86 gives you source, OTOH). Unfortunately, if you go that route you'll find it a lot harder to *sell* your software---it's the privilege of free-riding on Bill's immensely popular API that you are paying for in those commercial compilers, after all (along with customer support and other things, of course). -- Steve