Douglas, You seem to have misunderstood the idea of free, open source software, and what that implies. The authors have far more limited resources, especially for testing, than commercial developers. The produce software which you can try to use if you want to - they give no guarentees. They also give no guarentees that information (such as "this port should run fine on windows") are strictly accurate. For the majority of people doing serious development work, "windows" means something in the NT line. It is what the msp430-gcc group use (when they use windows), and it is what most users use. If you look pick ten random major open source projects that have windows ports, you will find that nine of them refer almost entirely to NT, and say that you might get it to work on Win9x if you are lucky.
Can we dispense with the innapropriate car analogies? Just to make it clear here - if you pay money for a car, you can expect it to work like a car or you can demand you money back - that is a fundamental consumer right (in most countries, anyway). If someone gives you a car and says "Here - try this if you want to. I can't be sure that it works, of course, but you are free to do what you want with it" then you have no rights for complaint unless you can prove that they deliberately sabotaged the car in an attempt to harm you. And the equivilent of "spilled the passengers onto the road" is "trashed your machine after installation" - this would have raised a lot more alarms than a "failed to run" problem. Things would be different if someone were to burn the tools onto a CD, and package it along with a simple jtag debugger and a printed copy of the manual, and sell the package for $100 (perhaps an idea for the msp430 gcc developers?). Then you have a right to expect the software to work (although not necessarily bug-free), despite the warenty on the gcc code. Do you understand why people have been reacting negatively to your posts? I can only really speak for myself, but I think others share these thoughts (and I expect to be flamed if that is not the case). It's not because no one cares about whether msp430-gcc runs on Win9x or not (we would all like it to run well on as many platforms as possible - but Win9x has very few potential users, and takes a disproportionate amount of testing and development effort). It's the implication that the developers are responsible for the problems some people have running msp430-gcc on Win9x, and that they are obliged to fix it, and that you can merely wait for new version to be posted. Open source software is much more of a two-way process. Lots of users are not interested or able to debug or fix the source code (although if you complain enough, someone will tell you to do exactly that), but there are lots of other ways you can volunteer to help. Providing information about what worked and what did not work (such as you have done now, regarding the two builds), is a big help, and offering to help testing is another. If you want to get the latest version of msp430-gcc to work on Win9x, then ask how you can help - don't just complain. David > You miss the point entirely.....The issue is continuing to pretend that 03 > 06 is a solution for the WIN98 or WINME platforms wastes time. The answer > was to use an earlier version 02 10 that doesn't have the compatibility > issue and at the same time alert others not to waste their time trying out > 03 06 for WIN98 and WinME. > I corresponded with an author who wrote a project article in Circuit Cellar > and he too was unaware if the issue so sadly many readers assumed 03 06 was > viable for Win98 or WinMe. The article went to press and much time was > wasted by readers trying this out on win98 winME. Some will want to fix 03 > 06 some will not, others will see it as a lost cause, but nobody wants to be > misled. > Now as to warranties misrepresentation is not a warranty issue. You can't > give a 4 seater Ford away with a back seat that spills the passengers onto > the road and avoid culpability by saying it wasn't warranted. Nor can you > say you ought to have bought a Chevy .If your Ford doesn't carry passengers > you have to rip the seats out to avoid culpability. > > ----- Original Message ----- > From: "Peter Jansen" <[email protected]> > To: <[email protected]> > Sent: Thursday, May 01, 2003 3:48 AM > Subject: Re: [Mspgcc-users] Win98 Segmentation fault > > > > I have seen this posting about the segmentation faults a few times now. > > I think the answer is if you have the problem and can reproduce it then > > fix it. The source for all the bits are available so figure out what is > > wrong and post a patch to which ever piece of code is causing the problem. > > > > If you don't want to fix it and suggest someone else should who has no > > interest in fixing problems on a system they are not using then I think > > you will be waiting a long time. > > > > If you don't want to fix it then do what is suggested and use a platform > > that others are using and is known to work. > > > > Its a free open source project, if you have a problem fix the problem > > and sent the patch to the project owner. If you read the warranty it > > states that the product is not fit for any purpose but may be useful, > > well if you want to use it in Windows 98 I guess its not useful. > > > > I don't want to look at the problem as I don't run windows 98 (because > > its unstable and Win2k/Linux are fine). > > > > Regards, > > > > -- > > Peter Jansen > > STS > > Australian Antarctic Division > > Channel Highway > > Kingston > > TAS 7050 > > AUSTRALIA > > Ph (03) 62 323 533 > > Fax (03) 62 323 351
