Don Armstrong <[EMAIL PROTECTED]> wrote: > On Thu, 03 Mar 2005, Matthew Garrett wrote: >> If we apply this to a photograph of a circuit board, we find that >> the photograph is the source. > > Quite possibly not, actually. Consider a > 2 layer PCB, FE.
Oh, sorry - I meant to go somewhere with that example, didn't and then left it in a horribly confusing state. I meant that a photograph of a circuit board that isn't intended to provide information about the layout of the circuit board should be acceptable as source, which might not be true if it's supposed to be a reference for board design. >> A 20 megabyte binary-only application is non-free, even if the >> author wrote and maintains it in a hex-editor. The author's >> preferred form for modification is a good metric, but not the be-all >> and end-all of whether a work provides sufficient freedom. > > Why not? Why must a work be in a form that you prefer when the author > finds it ideal for their work? What makes your prefered form of > modification special over the author's? I don't think /my/ preferred form of modification is more special than the author's, but if nobody but the author is in a reasonable position to alter the code then I don't think that's free. Free software is supposed to give us independence from the author - that's not possible if the work is effectively unmodifiable by anyone else. > The whole point of requiring sourcecode, as I see it, is so that users > (and Debian) have the same form that the author uses to modify the > code, so we're capable of making the same kind of modifications as the > author. I'd disagree - I think we want sourcecode because we want to be able to modify the work. That's subtly different to what you're suggesting, and there are works that could fall in one and not the other. From the point of view of modifiability, I don't think the author should be considered special. > Granted, I personally wouldn't package a work that was maintained in a > binary only form using a hex-editor for Debian, if for no other reason > than the fact that *I* can't modify the thing or audit it to satisfy a > reasonable level of quality. But that's not to say that Gods or > Goddesses of machine code can't package the thing. Mm. As I said before, I think I have a fundamentally different take on why we want source code to the general view here. Beyond what I've said already, I don't have a desperately strong argument for why mine is better. -- Matthew Garrett | [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
