Ubuntu's Unity Desktop: A Free Republic or Paternalistic Democracy?

If your computer is so old that it was last spotted in the wild roaming with the dinosaurs before they were flamed by an extinction-level event, then (like me) you just might just be grateful for Unity (2D) to extend the lifetime of your machine. Since the doctors switched off the life support on my best, though ageing laptop (private funeral only, no flowers, donations in lieu) I've had to switch the hard drive into my second best machine. The problem is that it's even older, at seven or eight years (probably about sixty eight in dog years). The spec' is miserable: slow single core processor, USB 1.0 ports, less than 700MBs of SDRAM and although it does support OpenGL there's not really enough horsepower under the hood to be really properly usable. My graphics card is not sub prime but everything else is. Therefore, "insufficient" for Unity-3D, the shell for Gnome slated for the imminent release with Ubuntu 11.04, codename Natty Narwhal and now the default desktop on Oneiric Ocelot (11.10). Dead end? No. Unity-2D to the rescue. Perhaps.

It's in, it's out. What?

The only good news on the horizon is that the latest release of Ubuntu will cater for those whose graphics cards won't support Unity. If you install/upgrade Natty Narwhal or Oneiric Ocelot and your graphics card isn't up to the task it will automatically fall back into a classic Gnome desktop. However, the bad news is that this sensible strategy didn't last. Earlier this year Canonical released Natty Narwhal's successor, Oneiric Ocelot and Mark Shuttleworth announced that the fall back would not be included. The original plan was to include Unity-2D with Natty but it was rather on the proverbial back burner and would only see the light of day when Ocelot took its first steps out of the jungle when it was proposed to add it as a third session option at log in. If you can't wait that long and your cash stash has been rifled and diverted to bail out those poor bankers (well, you want them begging on the streets, would you?) so that you can't afford to buy a computer which will support Unity then just thank your lucky stars that GNU/Linux is not Windows and you don't have to endure the expensive escalator of hardware upgrades to fill the already bloated coffers of Microsoft (no jokes please about how Canonical's benign dictator for life's initials are MS).

The problem is in fact a problem with my puffing and wheezing laptop, the East German Trabant of computers

Despite that, Mark Shuttleworth's recent announcements about Unity, Gnome and the future of Canonical and Ubuntu have caused some very real unease in the FOSS community. Many reviews of the beta of Natty Narwhal were hostile to say the least and now that oneiric Ocelot (11.10) is out the torrent of criticism (or should I say, abuse) has been nearly universally unfavourable, some going as far as to describe it as the worst Ubuntu. Ever. Reading them, I was forcibly reminded of the hell that broke loose when KDE 4.0 first emerged into the light. Well, it was buggy and looked liked it had been rushed out of the door prematurely. The central problem was with bugs and stability but after that rocky start it has improved considerably and I've started to use it again. Well, a bit. The problem with it is in fact a problem with my puffing and wheezing laptop, the East German Trabant of computers.

What we've got here is a failure to communicate

Only time will tell if Unity will improve and evolve and prove to be the smart move. Even if it is, it won't stop naysayers accusing Canonical of being anti open source, developing in house and contributing little or no code upstream. Bruce Byfield thinks it's the end of the honeymoon because the replacement of init with Upstart, Xorg with Wayland and Gnome Desktop with Unity was "politically" motivated and "not [by] the technical merits of the applications, but its ability to dominate the projects that dominate its software stack". Strong stuff. Yet most distros are now using Upstart, in-house development is hardly a conspiracy provided FOSS tools are being used and Wayland is a third party project which is being for adoption not only in Ubuntu but also in Fedora and MeeGo.

More recently, Byfield's blog on Linux Magazine this month includes a piece recounting an online exchange in Launchpad between Mark Shuttleworth and a Ubuntu loyalist which makes for interesting, and unsettling, reading; including a claim that Canonical tested Unity usability using fifteen Windows and OSX users. Is this actually true?

A conflict between purity and commercialism

My religious credentials may be zero but I have long suspected that there is a whiff of the misanthropic puritan about me, one that has always secretly lusted after the unsullied and pristine purity of edenic myth. When it comes to computing I feel it's gravitational pull too, so I am cursed with a Janus-like ability to see both merits of the purism of FOSS and the more commercially-minded approach of Ubuntu. I think Byfield's article has merit. To a point. But we live in a world where, unless you are a bank, you are never too big to fail and Canonical has yet to be profitable. That may not bother digital purists but its continued existence is important in order to maintain the corporate profile of GNU/Linux in general and Ubuntu in particular.

This matters because, despite any and all valid criticisms, Ubuntu has probably been more responsible than any other distro for raising the media profile of GNU/Linux and making it newbie friendly--and I say that knowing how good Fedora and Mandriva have been (and even better if they would use apt-get) but they have never got the same coverage in the mainstream press, the kind of coverage that draws in new users.

You still have choice (1)

I like opportunistic magpies, people who see an opening and take something in a different and entirely unexpected direction. The developers of Facebook and Twitter could never have envisaged the uses to which they have been put, including globally significant events. Obviously, Unity won't be a game changer but I do forsee not only a host of Ubuntu respins made by those who don't like where Canonical and Mark Shuttleworth are taking Ubuntu but people developing niche uses for Unity in other desktops. The choice of Unity for the name of the next Gnome shell is ironic as it seems to have generated a lot of disunity but one man's disunity is another's diversity and I think we will see a deal of it in the coming months.

Shuttleworth swore on his mother's grave that Debian was the rock on which Ubuntu would be built

Shuttleworth swore on his mother's grave that Debian was the rock on which Ubuntu would be built. I heartily approve of that. Let's keep it that way and not get too dazzled by the new, shiny things; by 3D effects and other eye candy. My experience has been that heavy and serious users of GNU/Linux tend to switch off or disable the superfluous eye candy in order to get any real work done. It gets in the way. When I'm researching and writing articles or tinkering with configuration files or settings I find invariably that I'm using the simplest, fastest and bloat-free desktops like Fluxbox or Openbox to simply gets things done. Fast. Without distractions, clutter, systems glitches and fewer hangs or freezes.

The beauty about Unity though is that, paradoxically, it has managed to kill two birds with one stone by allowing me to use fast, no frills desktops like Fluxbox and Openbox with the Unity launcher overlay to leverage a little extra functionality from an ageing laptop. (I'm still mesmerized by the sight of an "empty" screen with the Unity launcher down the left-hand side.) It's the perfect synergy of reducing, reusing and recycling. That was never really what they intended for Unity but I like to chalk that up as a little victory, under the heading: the law of unintended consequences.

The GNU/Linux universe is not entropic after all. It grows and diversifies and constantly reinvents itself

The GNU/Linux universe is not entropic after all. It grows and diversifies and constantly reinvents itself. One man's descent from order to disorder is another's man's luxuriant timeline resulting in multiple, parallel universes where both bloating eye candy and abstemious command lines co-exist happily together. However, desktop interfaces are a double-edged sword. Simple interfaces mimicking tablets like Android or the Apple iPad are a potentially good way to draw in users to Ubuntu but the problem is also, to use Richard Stallman's phrase, that when we use such interfaces we see through a glass darkly.

Stop taking the tablets

To paraphrase Stallman again, tablets never made jail so cool. Having tried out one myself I can testify to their seductive charms

When it comes to devices like tablets the glass is very dark indeed. These devices are for consumption, not creation (unless you hack them and install Ubuntu instead of Android they are not designed or fit for serious work). To paraphrase Stallman again, tablets never made jail so cool. Having tried out one myself in a local electronics store I can testify to their seductive charms. So can everyone else, including hardware vendors who see tablets and smartphones as the future. The PC is in danger of becoming a niche product but paradoxically its threatened demise could actually be an opportunity for Ubuntu (and others) to cater to power users, to fill the vacuum left by the mass adoption of tablets. And what does Mark Shuttleworth do? He writes an article on his blog saying that Ubuntu will be on tablets and smartphones--but not until 2014.

That's three years away and while a week is a long time in politics, three years is an eternity in the world of computing. By the time Ubuntu gets on these devices the curve will have advanced hugely and Android and iOS will have further entrenched themselves. Don't believe me? Just watch what happens to Microsoft's attempt to muscle in on the tablet market. Another part of the problem is that the percentage of people accessing the internet using PCs is shrinking and may decline to a demographic rump consisting of older users while most young people access the net on smartphones and tablets. That has potentially grim consequences for software developers who, increasingly, are having to reach their end users via OS vendors who filter apps and take a cut. This is Tim Berners-Lee's silos and walled gardens and it could kill the PC or reduce it to a niche product only for nerds and geeks.

You still have choice (2)

Ultimately, everything you do in a GUI is a convenient layer sitting over the command line and Unity threatens to take the end user even further away from the main reasons for using GNU/Linux in the first place: ownership and control. A Windows or Mac user migrating to GNU/Linux might just feel right at home but there would be little incentive to delve further. Until thing go wrong that is, and not being marinaded in the command line before distros became more "user friendly" they may simply give up and go back to closed operating systems.

Unity has obscured a lot of customisable functionality

With Unity the argument is not so much a toss up between transparent CLIs and opaque GUIs but about a GUI that has obscured a lot of customizable functionality. That won't be a problem for current users of Ubuntu who can and know how to jump ship to other distros or install other desktop environments and third party tools to tweak Unity. For Windows/Apple users it may just be yet another barrier to migration. The great thing however, is that the internet is awash with innumerable tips, hacks and howtos on making Unity more usable by incorporating features from previous Ubuntu incarnations. After all, I'm writing this article in Gedit on the minimal Fluxbox DE and when I reboot next time I might just take Enlightenment for a spin. Or Openbox, or XFCE, or LXDE, or Lubuntu or.....whatever suits my needs or whim. I'm not the only one.

Even GNU/Linux's benign dictator for life, the flying Finn, Linus Torvalds, has jumped ship. Again. First it was KDE, until the 4.0 release precipitated a flight to Gnome and then Unity pushed him to Gnome Shell 3.2 which he eventually warmed to on his Google Plus page, describing it as "almost usuable" after installing the gnome-tweak-tool and dock extension. This, after he had originally described it as an unholy mess and crazy crap and decamped to XFCE. In short he, like many other seasoned GNU/Linux users, won't be abandoning it anytime soon but Ubuntu's days as top dog may be numbered. When you try to be all things to all men you sometimes end up satisfying no one.

Unity in, synaptic out

While intense flame wars rage over Unity, Canonical has quietly dropped Synaptic as the default apt-get GUI package manager (though you can still install it after a distro upgrade because it is still in the repositories--but how many newcomers will know about that?). The Ubuntu Software Centre is a no brainer yes, but when something goes seriously wrong you often need a CLI or the configurability of a reliable stalwart like Synaptic. Fixing broken packages or pinning an app? What's that? Use synaptic? What's that guv? Newbies will have a lot of questions when things go wrong (and they will) but they will be chastened and possibly flamed on fora when they ask questions. That's the real subject of a flame war. I now find myself wondering how long Aptitude will last as part of the default software bundle too.

Democratising the republic of the desktop requires advanced and active citizenship skills. Ubuntu threatens to reduce users from citizens to mere subjects

The sad truth of the matter is that there is a limit to how much you can simplify the desktop without endangering usability, transparency or control. Democratizing the republic of the desktop requires advanced and active citizenship skills. Ubuntu threatens to reduce users from citizens to mere subjects--but just like voters, you can change your vote, abstain or even renounce you citizenship and become a free, independent rootless cosmopolitan owing your (temporary, conditional) allegiance to whatever you like. Like GNU/Linux, choice is free. And freedom.


This work is licensed under a Creative Commons Attribution 3.0 Unported License.