On Mon, 2003-03-10 at 16:50, Glenn Maynard wrote: > On Mon, Mar 10, 2003 at 02:36:51PM -0500, David Turner wrote: > > Indeed, in the current version, it is *perfectly clear* that mere > > modification triggers (2)(a) and (2)(c). If it did not, why would > > (2)(b) specifically mention distribution? > > Even if it's agreed that the current language restricts modifications > that aren't distributed[1], it's far from clear whether this was the > intent, or that it's useful. What's the point? It seems like a restriction > that has no benefit to freedom at all. Why do I need to date changes > for a program I'm not distributing? > > Of course, if I make changes and don't date them, I might have trouble > later on if I change my mind and want to distribute them; but that'd be > my own fault. The license certainly can't protect me from my own laziness.
The intent is actually to protect downstream people from your mistakes. Consider: person A at corporation X changes something in program P, then quits, and then person B prepares P for distribution, without knowledge of when A made what changes. If the changes are noted when they're made, this can't happen. > [1] The fact that there's active debate over this should be proof enough that > it's not "perfectly clear". Why not get an official position on this, don > the sombrero and settle it, so we can at least stop debating the wording? OK, I've asked upstream. Waiting for a response. -- -Dave Turner Stalk Me: 617 441 0668 "On matters of style, swim with the current, on matters of principle, stand like a rock." -Thomas Jefferson
