The conventional wisdom these days is that GNOME is faltering. GNOME 3 is unpopular, and users and distributions are abandoning it for alternatives such as Xfce or Mate.
The project itself suffers from a lack of developers and a loss of morale, and faces new challenges as mobile devices become more common than traditional desktop environments.
So what strategies are available for GNOME in the next few years?
This ugly assessment of GNOME's current condition is not just being made by outsiders. Recently, GNOME developer Benjamin Otte made the same critique in a widely discussed blog post entitled "Staring into the Abyss."
Many of the same subjects were even raised at GUADEC, GNOME's annual conference. In particular, Xan Lopez and Juan Jose Sanchez gave a presentation called "A Bright Future for GNOME" that outlined the project's challenges. Lopez and Sanchez's presentation was supposed to be a call to arms, but, in the weeks since it was delivered, it has been used mainly as proof of just how the once mighty GNOME has fallen.
Some GNOME developers have shrugged off the current malaise, noting that the project has faced challenges before. All the same, you can hardly help feeling sorry for members of the project. After all, many gave their time, labor, and ingenuity in an effort to create something new on the desktop.
But, instead of having their work accepted, they have seen it dismissed with anger and sarcasm. Some have also endured personal attacks on their competence and intelligence.
And now, after eighteen months of mostly keeping silent -- either out of project policy or simply out of the faint hope that familiarity would make users more accepting -- the GNOME team is facing the unavoidable conclusion that they have failed. It may be a necessary conclusion, but reaching it can't be easy.
So how could GNOME scramble out of its present situation? At least seven possibilities exist, although not all are equally likely. Some would even allow GNOME to back out with some grace from the dead end it's now in while remaining reasonably honest in the marketing strategy.
1. Modify GNOME 3
GNOME 3 was marketed as elegant, uncluttered, and easy. To describe it as a new vision for the desktop is not hyperbole, but a literal description of its designer's intentions.
With this perspective, the project has resisted making changes that might detract from the vision. The GNOME developers believed they had usability expertise on their side, and have shown little interest in offering more than one way to work in GNOME 3.
Yet with very few changes, GNOME 3 could be much more acceptable to most users. A moveable panel, panel applets, desktop launchers, user control of virtual desktops, menu alternatives that would remove the need for the overview -- all of these could be added easily as options. Together, they would reduce at least ninety percent of the complaints against GNOME 3.
Moreover, this way, GNOME 3 wouldn't be abandoned. It would just be modified to increase user choice -- a direction that would receive few, if any complaints.
2. Emphasize the Fallback Mode
The first releases of GNOME 3 required 3D hardware acceleration. Since many Linux machines still don't have hardware acceleration, they included a fallback mode that was a slightly crippled cousin of the last GNOME 2 releases.
In the last couple of releases, the need for hardware acceleration has eased, but the fallback mode could still be brought up to the standard of GNOME 2 in a couple of months.
The marketing strategy? GNOME 3 was a testing of two different desktop strategies, and the fallback mode has won. Users might even pretend to believe the claim, so long as they got what they wanted.
3. Ship with Cinnamon Extensions
Cinnamon is a series of extensions developed by Linux Mint that make GNOME 3 look and act like GNOME 2. It is also a major reason for the popularity of Linux Mint in the past year.
One advantage of Cinnamon is that the extensions can be added individually. If, for example, you want to add a bottom panel to the desktop, you can do so while keeping other features of GNOME 3. Alternatively, if you prefer not to have an additional bottom panel, you can turn off that option with a couple of mouse clicks.
Given this modularity, GNOME could save face by presenting the use of Cinnamon as an extension of customization options -- as giving users what they want. However, adding Cinnamon would mean that GNOME was using tools developed in response to GNOME 3's deficiencies, which would be a severe blow to the development team's pride.
4. Focus GNOME 3 on Mobile Devices and Cloud Services
If GNOME is having trouble as a desktop environment, one obvious solution is to find new niches. Lopez and Sanchez suggested following KDE's lead and producing a tablet, while Lionel Dricot recently suggested a suite of cloud-based services.
Superficially, both these ideas make sense. They take into account current computing trends, and GNOME 3 is far less annoying on a small screen than on a workstation, as anyone who has tried it using virtualization can attest.
However, the closer you look, the less practical either becomes. Moving into manufacturing is full of obstacles, as KDE is apparently finding out now with its Vivaldi tablet, which is now several months late.
Even more importantly, 2012 is two years late for considering a tablet, and four years late for cloud services. The markets for both are crowded, and to be a success GNOME would need some compelling benefits besides offering free-licensed software. Moreover, in the year or two that GNOME would need to produce either, who knows what the new trendy niches might be?
Both are areas that GNOME might want to explore, but counting on them to restore the project's prominence seems a gamble that has little chance of succeeding.
5. Keep Going and Pretend Nothing Is Wrong
In theory, GNOME could continue to be stubborn, and make plans without learning anything from the reception of GNOME 3. In public, its developers did manage to ignore criticisms for nearly eighteen months.
In practice, though, being stubborn could have a debilitating effect. To all appearance, it is the toil that such stubbornness takes that has depressed the project's morale and levels of participation. In the end, stubbornness would only continue the decline.
6. Limp Along And Quietly Retire GNOME 3
After GUADEC, the unofficial plan is to release GNOME 4 in 2014. Exactly what GNOME 4 might consist of hasn't been announced yet, but a major release would be an obvious time to back away from GNOME 3's design. Yet how does the project get there from here? What guarantee is there that GNOME 4 will be received any more favorably than GNOME 3?
The trouble is, this plan means another eighteen months -- if not more -- of the widely-panned GNOME 3. If morale and recruitment are low in the project now, what will they be like by the time GNOME 4 is ready? You have to wonder how many developers will be left.
Just as important, one of the problems that Lopez and Sanchez identify in their presentation is a lack of direction. If you lack direction to start with, the odds are low that you will easily find one.
True, you can cobble together a few goals and find some rationales for putting them together. Cynics might say that this is what happened with GNOME 3. Seeing KDE beginning a new release series, GNOME developers felt pressured to produce their own next major release, but had little sense of what it might consist. The risk is that GNOME 4 will be developed in the same way with exactly the same results.
7. Ask Users What They Want
The one strategy that GNOME has never tried is asking users what they want. Instead, the project has preferred to rely on usability theory, treating it as an exact science instead of a collection of competing ideas supported by usually inconclusive studies that could be mustered to support almost any design.
In GNOME 3, testing with actual users did not occur until near the end of the development cycle, when the chances of any major changes were remote.
Based on this performance, the chances seem slim that GNOME will suddenly start consulting users. More likely, such a suggestion would be greeted with explanations that users didn't know what they want, or what the implications of their demands might be. Others might argue that, if you only listen to users, then you will never innovate, because users only want what they already know.
A desktop -- and even a table and cloud services -- could be built by starting with what users know and adding new possibilities to it. But does the GNOME project have the will to work that way? Long-ingrained habits might make this approach impossible for GNOME to implement.
The Likeliest Strategy
Which of these strategies -- or, possibly, which combination of strategies -- GNOME will choose is still uncertain. Much depends on the ability of developers to admit that their efforts were wasted or unwanted, and to learn from GNOME 3.
However, the earliest indicators aren't promising. Having remade the GNOME desktop, developers are now intent on doing the same for the Nautilus file-manager, as The Blog of Helios points out.
The roadmap for Nautilus over the next year includes plans to remove the extra pane, and to merge functions that until now have been kept separate for convenience. With Linux Mint already forking Nautilus, all the pieces seem in place for a mini-replay of GNOME 3. Since a similar strategy produced GNOME 3, this choice seems dubious at best, especially since it would allow the project to continue.
Admittedly, the plans to overhaul Nautilus were being carried out before GUADEC and its talk of future plans and renewal. But the fact that existing plans have not been allowed to continue could be an indication that GNOME developers do not have the collective will to make the adjustments needed to revitalize the project. They definitely seem unable to change directions in a hurry.
In the short term, anyway, GNOME seems to favor the idea of limping along and praying for a miracle with GNOME 4. This strategy would allow the project to continue much as it has always done, but hardly seems an approach that allows for the broad renewal that is needed.
Probably, at some point, something called GNOME 4 will be released. But if the early indications are accurate, by the time it appears, nobody will be left to care.