One of the more interesting aspects of open source projects, to me, are the extent to which they seem to behave in ways considerably different than commercial entities. The open source world is frequently compared to the wild west, or other unsettled frontiers, while the alternatives are considered staid verging on boring. Whether its on mailing lists, Planet-aggregated blogs, or user forums, the range of interactions within open source projects is all over the map, from infantile name-calling all the way up to mature, reasoned and at least somewhat dispassionate arguments.
I hope you all noted the inclusion of the word “seem” in the first sentence above, because it’s an important modifier. All is not truly as it, well, seems. I personally would not make the statement that commercial companies are intrinsically more mature than open source projects; anyone who’s worked in one knows that the people and politics can be just as infantile and childish as the worst of what you get in the open source world, it’s simply that the open source world’s squabbles are a matter of public record, while the commercial world’s often, but not always, are not.
I bring all of this up b/c this past weekend saw something of a family squabble erupt within the GNOME community around notifications, design goals, etc, but as John Palmieri (J5) says, everyone hugged and made up at the end of the weekend. Like any functional relationship, whether its on a group or individual basis, the open source collaborative process will feature its share of disagreements, fights, conflicts, spats, etc. That’s not the issue, no matter how much critics of open source behavior might want to make it so; what is important is how they are handled. Can the various participants reach consensus or at least agree to disagree, and move forward? Such is the defining characteristic – IMO – of some of the best open source projects going, as GNOME demonstrated in this case. So kudos to the GNOME gang for defying the “open source is the mob” stereotype, at least on this occasion (can’t wait for a return to the language battles ;).
As an aside to one of the posts that I read: it’s certainly not my place to stick my nose into whether or not the netapplet/NetworkManager issue qualifies as a debacle or not, but I can tell you that as a GNOME user, I was not aware of the following:
Netapplet is a hack at best that more or less simply calls ifup and ifdown on interfaces. Its 3700 lines of code. We never intended it to be a GNOME solution or even a cross-distribution one. That wasnt because of any perceived value-add in it, it was because it was a stop-gap measure to fix the problem of switching between wired and wireless networks in SUSE, given tight restraints on development time.
The lack of clarity in the role of these projects actually led me to start this thread over at Gentoo, write this post, and drop a few emails. I’m certainly not questioning the need for throwaway or stopgap code – I’m pragmatic if nothing else – but as a user I’d like to know when that’s the case. I think every one of us in the technology issue tends to take things for granted (I take for granted all the time, for example, that people know who RedMonk is and what we do) so it’s certainly not an indictment of any the GNOME folks involved, but rather an indication that if the code can’t be perfect, more context about its limitations would be helpful.
