LINUX BLOG SAFARI

Ubuntu’s Game-Changing Quest for ‘Unity’

Human beings are creatures of habit, it seems safe to say, and tend to have a hard time embracing change.

So it has been no great surprise to see the widespread shock and dismay — mixed, still, with considerable excitement — over the decision to make Unity the next desktop Ubuntu’s default interface.

Yes, for those who missed it, Natty Narwhal — or Ubuntu 11.04 — will not use the distribution’s longtime GNOME Shell as the default desktop interface. Rather, it will feature the 3D and multitouch-enabled Unity, which just appeared in the netbook edition of Maverick Meerkat earlier this month.

‘I’d Rather Have GNOME”

Unity is clearly a big departure for desktop Ubuntu, and Mark Shuttleworth himself has acknowledged that the decision was a “risky” one. At the same time, it’s a clear indication that the distro is striking out in earnest and striving to make its own unique mark on the world.

Will the decision prove a good one? No one will really know until the distro comes out. In the meantime, Linux bloggers haven’t been shy about expressing their thoughts on the topic.

“I’d rather have GNOME,” wrote Quant in the comments on PCWorld, for example.

‘Moving in the Right Direction’

“Not so sure I like this,” agreed javaman90. “When it comes out, they had better make sure the old GNOME software works; or the Linux die-hards will jump ship.”

On the other hand: “Anything new but open-source is welcome!” countered ArnabGanguli. “I’m excited to give it a try.

“Besides, it’s not the only option to go with,” ArnabGanguli pointed out. “U don’t like it, snap it away and load gnome version…Ubuntu is moving in the right direction!”

Opinions on the topic seemed pretty widely divided on Slashdot as well, so Linux Girl did the only thing that made sense — she set off for the blogosphere’s gossip central, also known as the Punchy Penguin blogobar.

‘A Good Thing’

“Ubuntu has not dropped GNOME,” blogger Robert Pogson pointed out after a round of Tequila Tux Twisters. “They are just using a different default.

“For their purposes it works better,” Pogson explained. “FLOSS is a meritocracy, and it is the distro that determines the merit here. Good luck to them.”

Indeed, “competition between FOSS projects is a good thing,” noted Chris Travers, a Slashdot blogger who works on the LedgerSMB project. “It provides the potential for cross-fertilization of ideas.

“I see nothing wrong with Ubuntu trying to find something which they feel offers their users a better experience,” Travers added.

‘Likely to Promote Discussion’

Of course, “the market for window managers, graphical shells and the like is somewhat crowded,” Travers conceded. “It remains to be seen whether or not this is actually manageable long-term for Ubuntu.”

Even if it isn’t, though, “I think it is likely to promote discussion, which will lead to a better quality of experience for users of the Linux desktop.”

Slashdot blogger Barbara Hudson, who goes by “Tom” on the site, wasn’t so sure.

“Those who don’t study history are doomed to repeat it,” Hudson recited. “Remember the big flap over KDE 4.0? The first version of anything never works right. It’s why distros kept KDE 3.5 as the default KDE for several releases.”

‘Take Some Ritalin and Learn to Focus’

Changing to Unity, then, will “succeed in disaffecting their current users without offering anything compelling for new users,” Hudson predicted.

Instead, Ubuntu should “take some Ritalin and learn to focus,” she opined. “A while ago, it was netbooks. Then it was some silly announcements about moving into cloud computing. There was also the recent noise about ‘open core.'”

Not only that, but “after years of denying that their Hallowe’en Pumpkin color scheme was fugly, they finally dumped it for the slightly less offensive but still eye-challenging Iridescent Eggplant … and now they’re changing it again,” Hudson pointed out.

“I’m sure Red Hat and Novell, Mint linux and all Ubuntu’s other competitors will be very happy to see Ubuntu make the switch,” Hudson concluded.

‘They’ve Made MSFT Look Tame’

Slashdot blogger hairyfeet, however, wasn’t surprised to learn of the change in direction.

“Is anyone surprised?” he asked. “Frankly, watching GNOME I’ve felt they’ve made MSFT look tame when it comes to the amount of OSX ripping, and in GNOME it makes even less sense than Windows.

“In OSX everything is application-based and therefore having a central menu at the top and button layout like they have makes sense,” hairyfeet added. “Why can’t FOSS guys embrace something truly different, instead of either rips of the Windows or OSX desktops?”

‘My Hat’s Off to Canonical’

That, essentially, “is why my hat’s off to Canonical for going a different way,” hairyfeet asserted. “While I wish they would lose the top OSX style bar, or at least make it morph depending upon usage of application, watching videos of Unity I can say it is pretty impressive-looking at first glance and doesn’t appear to be aping either MSFT or Apple.”

Next, “if Canonical would just fork the kernel and make a truly dedicated linux desktop with hardware ABI, you might even get this old windows Greybeard to start stocking Ubuntu boxes again,” hairyfeet concluded.

5 Comments

  • What I said about GNOME was part of a larger conversation where I actually praised Canonical for being willing to try something new. Gnome seems to be obsessed with OSX, even where it makes NO sense, like having top level menus on an OS that is Windows and not application based. Anyone who has used OSX knows the two are NOT the same, and in the latter the top level menu makes NO sense.

    Likewise KDE seems more hooked on Windows, even getting more bling bling when Vista/7 came out. Surely the desktop metaphor can be improved, yes? Surely the same basic UI designed by PARC all those years ago can’t be the most efficient way to control a computer, yet time and time again I see FOSS rip off OSX or Windows. If you want Windows, run 7, the security and features are VERY nice. Likewise if you want OSX own a Macbook or build a Hackentosh.

    If Linux is truly gonna break out and get a real shot at general acceptance new ideas and new ways will have to be welcomed. Nobody is gonna go to the hassle of switching OSes for a ripoff. So my hat is off to Canonical for being brave enough to go in a new direction, and to try new ways of looking at the desktop.

    I only wish they would make a TRUE break away from the mainstream and fork the entire OS away from mainline. Linux has needed a stable hardware ABI for years and it is obvious that Linus and the rest of the kernel team only care about the server, so if Canonical truly wants a world class OSX or Windows 7 level desktop I truly believe that they will need to fork the kernel AWAY from Linus and the server devs, and with a stable ABI and kernel enhancements designed for desktop users, like priority for A/V tasks and desktop responsiveness, they can really light a fire under Linux adoption in the mainstream.

    • i agree with hairyfeet’s comment

      i would also like to see canonical do this and add support for hardware ABIs

      it would be tricky of course (what would be the easier way to doing it? adding support as a module? or actually getting their own branch?) and they might want to wait a bit till their desktop is a little more stable and their relationships with hardware manufacturers become stronger.

      However for this i think canonical needs first to either:

      -slow down "official" releases to one year

      -and/or create a rolling version (like mint debian)

      the rolling version could either be the "test or main" version (or both?)

      these would be the easier temporary fix to what hairyfeet is asking i believe (um, can i get an opinion on this?)

      • "I only wish they would make a TRUE break away from the mainstream"

        To be part of the mainstream as a PC, you have to look like a PC which means you have to look like Windows and the world is not kind to look-alikes.

        Android has a much better chance since people expect it to work like a phone, i.e. with the same "apps" as they may have become used to. If an ARM netbook is to succeed, it will have some form of Win7 on it if it is marketed as a PC. Ubuntu could start a campaign to sell the "cloud book" or some similar concept, I think, and could be the leader. They would have to come up with some sort of business plan that could pay for the promotions,though, and I don’t think that is very easy to do. Google can compete and sell apps for Droid, but Google is a special case, too. Microsoft, Google, and Apple are well-known brands and can do things that the no-names cannot do.

        • What my opinion on the matter is is thus: Ultimately ripping OSX or Windows will NOT gain share or users, especially like in the case of Gnome where ripping OSX makes NO sense.

          As for the ABI, I don’t think it would be tricky at all. Canonical is already running THE home user based Linux, and compared to the other desktop distros has the lion’s share of the "admittedly small" desktop Linux market. This gives them power, just look at how Shuttleworth was able to get many apps to conform to the Ubuntu release schedule.

          But as someone who builds and services PCs for a living I can tell you Linux has a MAJOR flaw, one that isn’t shared by the other OSes out there: The lack of a stable hardware ABI. Solaris, BSD, OSX, and Windows ALL have a stable ABI, which makes it trivial to "write once, use for years" with drivers. For example I have a customer who is using Win2K drivers for a piece of hardware on XP Sp3, and I myself AM using Vista 64 drivers for a few specialized pieces of gear on my Win 7 X64.

          As it is now you will have to recompile drivers for new kernels, which means either A-a company has to keep developers on hand to constantly rewrite drivers, ala Nvidia, or B-place the future of your devices on Linux in the hands of some kernel developer who may/may not care or have time to write constantly optimized drivers for your device.

          And of course BOTH of these routes means you simply cannot "put a little penguin on the box" and have the device work like you do with OSX and Windows, and this is simply unacceptable if Linux is to ever gain traction with the masses. So by writing a common kernel interface that NEVER changes Canonical could give device manufacturers a trivial way to support Linux and put a penguin on the box.

          This would also allow guys like me to sell and support Linux, as in the current situation the 6 month upgrades often break drivers, which means lifetime free support that would frankly break me. Unfortunately the "hardcore" faction led by RMS has made this a political issue, as having a stable ABI means a company could just write drivers without giving their code up, or depending on the mercy of a kernel dev for support.

          Anyway I hope this has enlightened you. With a stable ABI one could put penguins on boxes, which equals easier shopping for consumers and easier selling for shops like mine. Both of which would help spur adoption by making it trivial for consumers to buy and use Linux without any of this "open up bash and type" nonsense, which if you check the forums is a good 95% driver related. Until this situation changes you simply won’t see Linux gain mass traction, because it is still too much a "DIY hacker" OS, and that simply isn’t what the consumer wants nowadays.

          • I totally agree with your Idea and AM sure many more will agree too.

            We’ve reached a stage where it makes sense for canonical to do so, specially if it plans to get more support, mature, become a better competitor and grab market share.

            Not only you will reduce on support, but Canonical would be the first one in reducing these support issues, reduce costs, reduce complaints and even get better reviews.

            You should get your idea to the attention of the developers ASAP, they’re improving the feedback greatly. Check here:

            http://mdzlog.alcor.net/2010/11/03/weathering-the-ubuntu-brainstorm/

            it may not get implemented right away, but the sooner one starts to talk about it, get people to back this and get the developers attention, the faster we could see a solution.

Leave a Comment

Please sign in to post or reply to a comment. New users create a free account.

LinuxInsider Channels