On Thu, 2004-11-18 at 23:45, Hamish Moffatt wrote: > On Thu, Nov 18, 2004 at 04:59:36PM -0600, Manoj Srivastava wrote: > > On Thu, 18 Nov 2004 22:51:42 +1100, Hamish Moffatt <hamish@debian.org> said: > > > On Wed, Nov 17, 2004 at 09:37:01PM -0600, Manoj Srivastava wrote: > > >> -- If you only have a hammer, you tend to see every problem as a > > >> nail. Maslow > > > > > Your post above demonstrates this aptly. > > > > And how so, pray tell? What is the "hammer" I have that leads > > me to classify computer related entities as a) software, b) hardware, > > or c) wetware? > > The hammer is your extremely limited vocabulary for classifying > computer-related things. Suddenly you've classified everything as > software and lost any subtlety of documentation versus programs > versus data etc. "Suddenly you've classified everything as matter and lost any subtlety of plastic versus metal versus gasses etc." Just because I choose to call programs, documentation, and data all "software" doesn't mean we can't distinguish between them. It just means when I use the word "software" I mean all of them, not caring what particular kind of software it is. The same way when I say "all matter consists of protons and electrons" I don't care about the kind of matter. If I do want to distinguish, I use "program", "documentation", "data", etc. For example I might say "I think that Debian should have different requirements for freeness in documentation than in programs." But it doesn't make sense to say "different requirements... in documentation than in software", anymore than it makes sense to say "Titanium is a metal, not matter." One is a subset of the other. Some people didn't interpret "software" this way, apparently, which led to many flamewars about the editorial changes GR. If they had spoken up about their perceived problems when the GR was proposed rather than after it passed, things would have been easier... I also think, when you begin to examine questions of classification into program/documentation/data, the line becomes impossibly blurry at many points. Using "software" to refer to all of them avoids that problem entirely. If someone wants to assert something like my hypothetical statement above, I think they have to propose (or at least outline) such a distinction criterion. (ObBitter: But no one is interested in having this discussion, because it involves a lot more thinking than just repeating "documentation is not software!" or "the FDL is free!" or "no consensus!") -- Joe Wreschnig <piman@debian.org>
Attachment:
signature.asc
Description: This is a digitally signed message part