Conflicts of Interest: Plans for Nautilus and Evolution
Looking at the big picture
Even as GNOME 1.4 gets ready for its first beta this weekend, Miguel de Icaza was priming the pump for 2.0. with a document outlining the possible directions the project may take.
The issue, according to Miguel, is whether to maintain GNOME on its current trajectory, looking primarily to build on the established base later 1.x series releases have provided, or to "pull a KDE", go blue sky, and produce a 2.0 release that changes the project dramatically. In an interesting nod to the esteemed competition, the document mentions how a C++ version of bonobo might attract more developers and make bonobo a more commonly used tool. If we're lucky, enough howls will go up over this to drown out the more tedious droning from the desktop warriors. It's a sad day in Linuxland when you pine for a hearty C/C++ fistfight to relieve you of the monotony of the bleating from the sidelines. Next I'll be walking up to strangers and yelling "Emacs! Emacs! Emacs!" while I poke them in the chest.
The two big star apps of the GNOME 1.4 release are Nautilus, the new file manager from Eazel; and Evolution, the mail client/PIM. Falling in behind the apps (from an end-user perspective) are the GNOME-VFS libraries, which provide a way for GNOME apps to easily access virtual file systems (like you find in archive files of various sorts), and bonobo, the GNOME component libraries. It looks like Evolution is going to arrive later than the general GNOME 1.4 release by a notable stretch, even as Nautilus is scheduled to come in on March 5th, several weeks ahead of the rest of 1.4.
The real heartache for "Joe Casualobserver" in the march to 1.4 has centered around the fact that there's no easy way to see Evolution and Nautilus running alongside each other much of the time, largely owing to conflicting bonobo version dependencies, though the GNOME VFS libraries have seemed to cause a little heartburn here and there. An Eazel developer I spoke to recently promised to "help Ximian have a soft landing into GNOME 1.4" when they finally ship Evolution, which was an awfully friendly thing to say. I'm sure the folks at Ximian are looking forward to a helping hand.
Ximian and Eazel form an interesting pair in the GNOME world. Most people paying attention to the two companies are aware of some cultural differences between the two, which will make competition between the two businesses all the more interesting. Once they're done going along to get along to make a complete GNOME 1.4, there's a lot of room for strangeness with the companies' plans to make money on the new Open Source mantra: services, services, services.
Eazel has made no bones about their focus on quality testing and making deadlines. Observing Nautilus on and off since last Fall has been an interesting trick for anyone besides a Red Hat 6.2 user, since Eazel has (with the exception of recent allowances for Red Hat 7) been using Red Hat as a reference platform and largely eschewing any other distribution in terms of the binaries they'll provide. There are some good reasons to do this, that speak volumes about Eazel's pragmatism in entering the Linux development world. They've saved themselves a lot of headaches by developing against one primary distribution. The peanut gallery can trivialize the differences between distributions as much as they want, the people actually developing commercial apps (as opposed to those who consider knowing a few different menu schemes to be "cosmopolitan") frequently mention the hassles of trying to support even the last three point releases of the major distributions. It's a pain. Stormtroopers aren't coming for your computer because developers are saying that, so feel free to say it yourself: bunches of different standards make life hard on people who want to build software.
Eazel's narrowness of focus, however, is something that goes beyond which distribution it's built to run best on, and may even be extending to who gets to take advantage of the full range of services Eazel will be offering.
Several weeks ago, Eazel and Red Hat announced a partnership that will make Nautilus a conduit for the Red Hat Network, and a recent interview I had with a pair of executives from Eazel indicated that the services the company offers in the form of downloadable software may be very limited to users of distributions other than Red Hat. In fact, though they have plans to eventually integrate other popular distributions (like Mandrake) they aren't even sure where to begin to address offering services to Debian users. So, Red Hat's the distribution against which Nautilus is being most tested, and it's likely the distribution that will benefit most from Eazel's services.
A colleague recently remarked that gossip had Red Hat feeling resentful toward Ximian for "hijacking" GNOME, which called up several-year-old memories of the initial uproar over Red Hat's support of the fledgling desktop environment and cries of how it was a case of "NIH" syndrome blown to its most evil proportions in an attempt to destroy KDE and any distribution anchored around it by default. Ximian's hardly the issue, though. Eazel has built a mechanism that will provide not only another attempt to make the Linux desktop salable to a lot of consumers via easy software installation and other usability fixes, but a tool Red Hat can use to make a case for both its networking/corporate oriented services (by providing a conduit for the Red Hat Network), and whatever rear-guard action it hopes to maintain on the Linux desktop/consumer market (by pointing to the fact that the most whiz-bang neat features of the GNOME desktop will only happen on a Red Hat box.) In other words, if Red Hat ever thought of "controlling" GNOME, the current GNOME landscape does nothing to indicate they've got reason to be upset: they're the reference platform for the biggest, earliest-to-market "value add" to the whole thing: Nautilus.
So, that leaves us with that strangeness that was mentioned earlier.
The GNOME desktop is going to be a money-making conduit for two entities that both have plans around the same theme: selling stuff to people over the conduit that is GNOME.
Ximian doesn't control Nautilus, exactly, so Red Carpet (due for a first release next week) is going to be their software pipe to the end user. Eazel won't be doing much for the next few months besides getting its feet wet with the Red Hat userbase. It's not even clear where all the possible points of contention will be, with the only "definite" seeming to be the sale of software over the 'net.
I think the expectation is that columns like this are supposed to end with predictions of either blood and mayhem or sweetness and light, but this seems to be much more interesting as the start of a discussion over what happens when a pair of companies with competing interests decide to meet on a common platform and work to better it while angling for slices of each others' markets. I've got an optimist and a cynic, one for each shoulder.
'Name That Column' Contest Extended
I got a lot of entries for the 'Name the Column' contest I announced last week, with all sorts of people writing in to contend for the Ximian stuff (hat, t-shirt, monkey, CD). Thanks to some mail problems that bounced a lot of entries back, we're going one more week to wring the last drops of creativity from the audience. Mail me your suggestions, and a dedicated cadre of LinuxPlanet writers and editors will pick a winner.