On Sun, 13 Feb 2005 15:02:53 -0500 (EST), Stuart Brorson <[EMAIL PROTECTED]> wrote: > Remember that there are still lots of RH7.X installations out there > serving as desktop systems. [...] > I'll belive you, but please don't forget about the user! Users don't > like to be on the mandatory upgrade treadmill. Recall that that's one > of the charges against Microsoft: they put you on a treadmill.
An apples-to-oranges comparison, IMHO. The RH7.x users still get to keep their core OS. The Microsoft upgrade treadmill is different-- usually refers to replacing an entire OS, and needing to upgrade hardware as well as software. (I'm surprised that nobody has brought forth any statistics showing how much slower GTK+ 2.x is than 1.2, since it's been discussed to death in a number of other non-EDA projects. Admittedly it's not as bad as going from, say, NT4 to 2000 in the Win32 world, but still...) If you're really that concerned about simplifying things for the end user, what about static-linking the binaries, and just distributing the source along with them? The right to source code is a separate issue from making that source code useful for an end user (who doesn't necessarily know about building from source). End users have at least the following options when they consider a binary version of a tool: * Trust the person who compiled it, * find another binary distribution they can trust, * or find someone (whom they trust) to compile the tool from source. Programmers' time is limited. If I had Bill's patience to wade through converting all the Xaw/Xt stuff in PCB, I would want to take advantage of all the time-saving functions in the new GTK+ libraries, too. Finally, there's nothing that says that a user who is interested in using an older OS *must* upgrade to the latest-and-greatest gEDA, either. -- - Charles Lepple
