summaryrefslogtreecommitdiff
path: root/published/thefutureofdesktoplinux.txt
blob: a72edd4d8059db3433da178286727829221532d4 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
After such a <a href="http://www.theregister.co.uk/2014/12/22/linux_year_review/">banner year</a> of Linux releases it might seem overly pessimistic to pause and ask, <em>but is there a future beyond this?</em> 

The answer is of course yes. Or rather it's yes, *but*. The qualifying "but" can take many forms, depending on who you're talking to and what their stake in the game is. 

Even if you take the most optimistic outlook for the future of the Linux desktop -- that all these distros continue turning out all these great releases year after year... until what? One day there are no more laptops or desktops left?

In this sense the question of what the future looks like for the desktop Linux distro is closely tied to what the future of general purpose computing looks like.

And that future looks increasingly like one that will exist primarily on mobile devices.

The much-hyped mobile future may not feel that promising or real to those who grew up with TRS-80s in the den and will always think of computers as something you can hack on, something you control; but for the other 99.99% of computer users, mobile devices are in fact exactly what they want for the very same reason -- they never wanted to tinker with that massive, ugly contraption like that TRS-80 sitting in the den in the first place.

The tightly controlled, rarely compromised world of app stores and mobile contracts is not limiting if that's all you wanted in the first place. It may not be the best way to debug your company's failing mail server, but mobile devices are great for getting you search results, Facebook updates and simple ways to share photos with friends. 

Couple that with the reality that the next 2 billion people who will be connecting to the internet will be in areas where factors like price, battery life and portability make the mobile device a clear winner and you can see through the mobile hype to a more nuanced reality, one where the device that's good enough trumps everything else. The exact form of that device will change over time, but right now the phone form factor seems to be winning.

Given Linux's long history of outstanding support for underpowered hardware, mobile devices could end up being the best place for Linux yet. Unfortunately, so far the gap between "Yes! Mobile Linux will be awesome!" and an actual mobile version of Linux that runs on devices that actually sell in stores is, well, insurmountable.

The most visible face of Linux in mobile and, let's face it, the most likely to succeed beyond the small circle of the Linux faithful, is undoubtedly Canonical.

The company has been hard at work on Ubuntu for phones for some time. In fact, there has been, for the latter half of 2014 anyway, little to Ubuntu other than mobile. As Canonical's Jon Melamut, VP Professional and Engineering Services, tells The Register, "Ubuntu for phones has been a major focus for Canonical, and we’re now in a position to bring devices to market." Indeed it looks like we'll see the first official Ubuntu phones in the very near future.

Exciting as that is for those who've been waiting for the power of Linux to make its way to your hand, it's really only a halfway step to Canonical's vision of "true device convergence."  

It's still a ways off, but Melamut says Ubuntu for phones and Ubuntu desktop "ultimately... converge into a single, full operating system that will work across different form factors from mobile to tablet and PC." Interestingly, while many of the other big players initially scoffed at the idea of one OS to rule all your devices, that now seems to be exactly what Microsoft and Apple are now moving toward as well, albeit in very different ways.

Ubuntu may be pushing hard to get a slice of the mobile pie, but that doesn't mean every distro is likewise inclined. In fact, chasing mobile might be missing perhaps the biggest opportunity desktop Linux has ever had for widespread adoption. The masses may be swapping their aging Windows XP desktops for tablets, but the so-called power users are unlikely to do that now and won't be likely to do it in the future either. 

With Windows 8 proving unpopular with power users and OS X Yosemite eliciting pointed criticisms and even talk of Apple <a href="http://www.marco.org/2015/01/04/apple-lost-functional-high-ground">losing</a> some of its "just works" magic, the Linux desktop may well be the refuge power users and developers are looking for. Melamut points out that while Ubuntu for phones gets all the press, Canonical has also been ramping up its "cloud" options as well, targeting dev ops, cloud developers and the like with OpenStack environments and other tools that make developer's live easier. If you're using it in the server room, why not your desktop as well?

That's the user that the openSUSE project plans to focus on, chasing what openSUSE Senior Consultant Douglas DeMaio calls "the power users" and making openSUSE, in DeMaio's words, "the MIT of Linux distros". 

The focus on power users isn't new for openSUSE, which has long offered the very sysadmin-friendly YaST as one of its big selling points. Over the past couple of years openSUSE has completely re-written YaST in the more developer friendly Ruby language (previously YaST was written in a homegrown language) in an effort to draw more power users and sysadmins to YaST and the openSUSE platform.

DeMaio says that openSUSE plans to continue improving YaST and focus on other power user features and tools like the recent move to Btrfs as the default file system. For openSUSE the name of the game is evolution, not revolution. "The innovation is in the process," says DeMaio, "it's in the tools."

The Fedora Project has taken a similar approach, but in a more radical way. Fedora has long been a massive, sprawling project with dozens of different "spins", specialty deployments and niche packagings. On one hand this is part of what allows Fedora to bubble up some very interesting projects, like the <a href="http://devassistant.org/">DevAssistant tool</a> for developers or its robotics package setup. The problem with this nebulous approach is that the lack of cohesiveness has made it hard at times to figure out where the project is actually headed. It often seems like it's headed everywhere and nowhere at the same time.

That changed this year with Fedora.next, which essentially envisions Fedora as an onion with several different, but connected layers. There's a couple core components, a bare OS and then a base system -- Fedora minimal if you will -- then comes something new, what Fedora calls environments or, or if you prefer the marketing term, "flavors". These are the different setups that arrived when Fedora 21 was released. Right now there are flavors for cloud, server, and workstation environments (the latter is basically a desktop environment geared at software developers).

Fedora Project Lead Matthew Miller likens the new structure to Lego. "The idea is: we can take some of our bricks, and we can ship those as sets," says Miller, "and maybe even, unlike Lego, we will ship them preassembled for you, but we’re not gluing them together, and we’re not getting rid of the basic supply of bricks."

The modular approach also sets the stage for other directions in the future. Right now the flavors are cloud, server and workstation because those are the places the distro is focused. Down the road if Linux-friendly hardware emerges in the mobile device world then perhaps we'll see a new "mobile" flavor from Fedora.

Fedora's approach mirrors what's long been the guiding principle of good GNU/Linux software -- small parts loosely joined. This will no doubt cue comments about the evils of systemd, but at the structural level at least the modular approach seems alive and well in Fedora.

So far though that modular approach has not jumped on the mobile bandwagon.

While most of the Linux world may be ignoring the mobile future for now, or perhaps waiting to see how Canonical fares, there's a second, slightly less glamorous possibility for the future of the desktop PCs in which Linux fares quite well -- the return of the desktop/laptop. 

When the iPad first arrived pundits proclaimed the end of the PC and so forth, but in fact what many of us learned from dabbling in mobile is just how valuable our laptops really are. And when we get back, desktop Linux will be there waiting with open arms.