Tomas wrote:
Atleast they are trying, which is more than i can say about you whiners.. If everyone believed like you, then there would be no software..
Of course, shoot the messenger. Always a good tactic when you are lacking proper arguments. So let's keep the dicsussion which will now ensue polite and to the technical point, okay?
Look guys, even though I do not own an Amiga anymore, and do not intend to buy one either, I am not against it in any way. I'm sure there are still many people who enjoy the little computer and would love to have decent software for it. It is just that Mozilla is a bridge too far. Why? Because it uses software technology (hate that damn word, but okay) which is quite alien to the Amiga,
especially OS 3.1. It uses POSIX-compliant threads, which do not exist as yet on the Amiga, and are hard to implement too. It relies on the GTK+-toolkit, which does not exist at all on the Amiga. It uses the Boehm garbage collector, for which a sort-of-port exists for the Amiga, but operates under severe restrictions. I do not know how that will affect the Amizilla. It uses fork() at some places, the absolute Unix-to-Amiga porting nightmare. The dynamic library interface is completely different to what is common on the Amiga. You cannot even compile the sources on an Amiga because of their insane size running into hundreds of MBs. And so forth.
Writing decent implementations of each of the above parts are formidable and very difficult tasks in their own right, and now people are striving to do all of them at once! With AmigaOS 4, POSIX threads and fork() may be properly implementable at last, but despite my asking on several occasions on these forums, the Frieden brothers have not yet confirmed this. In any case, the impression I got while studying the sources was that you need to create so many extensions to the AmigaOS to get Mozilla going, that you might as well offer your services to Hyperion (or Genesi) to aid them in coding their next generation operating system. And at a salary equivalent to the bounty money.
PeroxideChicken, you hit the nail on the head. The problem is
exactly that Mozilla did not start out on the Amiga, but was created and designed on a more modern platform. It makes perfect sense to use code that is already available for that platform. If such code does not exist on a much older platform, you're out of luck. On the other hand, Scala and Lightware originated on the older platform, and thus can be much more easily ported to modern ones. The software can even be adapted to take advantage of the newer infrastructure in ways which would be very difficult if not impossible on the old one.
Bottom line: I am
not negative, but I
am brutally realistic. There's a difference. Unless people can convince me that the above subprojects are easily resolved, I will maintain my opinion that Amizilla is dead in the water. Please keep in mind that I am not saying that people should not do anything at all: there is indeed a lot you can learn from the process of porting a beast like Mozilla. Let them thinker to their heart's content: perhaps it will aid them in porting other programs. However, if you are contributing code with the bounty in mind, your efforts are much, much better spent elsewhere. The same goes for people who donate money in the hope that it will result in a working program: the technical issues are such that the bounty needs to gain at least another two digits before the effort becomes worthwhile. I am very sorry, but that's the way it is.
