How does Linux need to change to achieve the 'year of the Linux desktop?'

Also in today’s open source roundup: Ubuntu 17.04 will be called “Zesty Zapus,” and Ars Technica reviews the Google Pixel phone

Changes and the year of the Linux desktop

The year of the Linux desktop has long been the unicorn of the open source movement. People have looked for it for quite a long time, but nobody’s ever seen it. But hope springs eternal, and for some the quest for the year of the Linux desktop goes on.

One redditor recently asked what needs to change in Linux for next year to become the “year of the Linux desktop.” He got some very interesting answers from his fellow Linux redditors:

0Oceanman0: “What do you think needs to change about GNU/Linux in order to make next year, the year of the "Linux Desktop.”

Eleventhousand: “I don’t think desktop computing excites many people anymore (unfortunately).

It’s still a little bit too inconvenient for laymen. For example, users are used to just pressing “install Windows updates now.” These same users will be lost in Ubuntu when it fails to install updates because there is no room due to a bunch of old kernels taking up space. Linux requires that a user run shell commands from time to time, which is a deal breaker for laymen.”

Panorambo: “I don’t think the two problems cancel each other out. You have the desktop problem, where fragmentation into different toolkits and applications linked with these, of differing quality and sets of features, divides users and basically hinders all kinds of good things happening. And then you have the lack of applications that allow people to manage their computer without resorting to command line. The two can be seen as part of the same problem of Linux on the desktops, but from the point of view of solving these, the solutions can be developed independently and thus in parallel.

The solution to the first is to define a set of protocols that can separate a GUI toolkit implementation from a mechanism. This allows developers to not have to make a choice about which toolkit to adopt for their application, instead targeting the protocol, which is used by users preferred toolkit at their end.

The solution to the second problem – irrespective of whether the first problem is solved or not – is to devise a way for a user to e.g. manage disk space without use of the terminal [emulator]. Yes, command line is immensely powerful, but disk space management? Seriously? Please don’t tell me it cannot be solved through a GUI? Are there other problems regular users are facing that should deny them functionality without the command line?

Sometimes I think the problem is psychological – many people use Linux because they like the command line. And so the circle is complete – there are not enough incentives to design great desktops (again, regardless if the fragmentation problem is solved or not), because generally everything can be done through the command line. Well, you know, if you are good enough with it.

P.S. It is ironic how another attempt at winning the user, Canonical’s Unity is called that. What does it unite, exactly? Not hating, but there is some irony there, is it not? Also, Mir means peace in Russian, is the idea behind Canonical’s window manager to, I don’t know, pacify the community?”

Cardboard-cutout: “Honestly? More games, more specialist software.

If I could play my games and use the software I needed I would be all over it.”

Remotefixonline: “Other than games and photoshop stuff for home users, the corporate software needs to have a Linux version. If corps could easily switch they would, but so much relies on ms exchange and sql these days.”

E_t: “As a user of Linux on the desktop for the past eight years, I’m pretty happy as things are. I think the biggest obstacles to wider adoption are external to Linux itself.”

Ecmdome: “Agreed. Adobe for Photoshop/Illustrator and with the right driver support Premier would be a great place to start.

Same with support for ProTools. I know that these communities would take well go Linux given the proper support.”

Panorambo: “It doesn’t bother you that developer communities are still split into at least KDE and GTK camps? Even if an entire suite of killer apps would arrive, like Photoshop, GUI applications on Linux are traditionally linked with either KDE or GTK, swaying user groups one way or another, without good reason.

I’ll be happy when we start thinking in terms of cross platform protocols, instead of implementation/libraries. So that we can have GUI applications instead of KDE/GTK-applications. Much what did with the window management, we could do with toolkits, accessibility features, etc. I know some of it is covered by Freedesktop, but it is either is not looked into sufficiently (aka “nobody cares”) or is not there yet. When protocols drive applications – definitely a higher level abstractions than dynamic linking to a particular implementation – at least developers don’t have to make the choice of writing either a GTK or a KDE app (as an example), leaving the choice to the users. This, in my opinion, will greatly improve the state of things on the desktop.

I don’t use Photoshop, by the way. But I still need to be aware whether the app I like is KDE or GTK. And having installed both, it is up to me to make sure the application blends in, which it seldom does, because well, there are not enough agreements about configuration schemas, looks, and behavior. Whenever I hear about people joking about the Year of Linux Desktop, I think of that deficiency, first and foremost.”

Gimpy1405: “I’m pretty sure next year won’t be the year of the Linux desktop, but Linux and open source keep growing so who knows? Desktop itself is becoming less used as people migrate to mobile devices. I wonder if there will ever be a year of any desktop.

But to answer your question, easier installs on more machines. Too many distros require more expertise than the average user has. Certainly more expertise than I have. About the only slam dunk installs I’ve had are from the extended Ubuntu family.”

Zzspectrez: “It already arrived and you missed it!

I haven’t used windows on my home machine for many years.

There will always be some proprietary app that you can find that is not avail. Photoshop, Autocad, Microsoft Office, Adobe Premier, etc…

But even then there are often alternatives..

If you are looking for a windows or mac experience then stick to windows or mac.

Sure things can be better, but everything I need to be able to do I already can.”

Bull500: “A good user selection process without the hassle

a. a website that allows you to chose the desktop env. you like after viewing a demo, 32/64bit and multiple download options - This exists but can be heavily improved for newbies

b. When you have a global rpm/deb file format for all linux’s. Click and install. It should be as simple as that.

c. A startup tutorial after install. Or official online ones for viewing later.

d. a proper explanation of why open source and where additional software’s or repo’s are located and how to install them(eg vlc)

e. Show how OSS can be compatible with proprietary software’s(eg LibreOffice). Or how the OSS ones are much better in comparison.

f. Better UI/UX, icons, themes.

g. Easier Proprietary driver support/installation”

Gg86: “Nothing, at this stage it’s just more popular apps, OEM buy-in and a marketing push. Ubuntu is my default OS, the one I fall back to unless I need something specific and that’s been ready for years now.”

More at Reddit

Ubuntu 17.04 will be called “Zesty Zapus”

It’s always fun to find out what the next version of Ubuntu will be dubbed by Canonical. Ubuntu 17.04 will be called “Zesty Sapus” according to a report on OMG Ubuntu.

Joey-Elijah Sneddon reports for OMG Ubuntu:

In place of a Zany Zebra or Zealot Zonket, the next version of Ubuntu is named: ‘Zesty Zapus’

Zesty is an adjective meaning ‘great enthusiasm and energy’. It can also mean that something tastes rather citrusy (as citrus fruit peel is called ‘zest’).

Zapus is the genus name of a North-American meadow jumping mouse. The cute little rodent is said to be the only mammal on Earth that has a total of…18 teeth. Everyone needs a claim to fame, right?

More at OMG Ubuntu

Ars Technica reviews the Google Pixel phone

Google’s new Pixel phone has certainly been noticed by many Android users. But is it worth buying? Ars Technica has a full review of the Google Pixel Android phone that should help you make a final decision if you're on the fence about buying the phone.

Ron Amadeo reports for Ars Technica:

As a rookie OEM, Google has produced a middling piece of hardware with the Pixel phones. The design is a bland iPhone rip-off with a clunky back panel bolted on. The large and empty bezels feel like a downgrade from past devices, and the phones also have a few objective downgrades, like price and speaker quality. In terms of price vs. performance, the Huawei Nexus 6P feels like something that was designed and built by an experienced, well-oiled machine, while the Pixel really does feel like a new company’s first attempt at hardware.

The hardware is also unambitious. For Google, becoming an OEM meant shaking up relationships with its Android hardware partners. Despite the company’s assurances that Samsung and friends are fine with Pixel, this was a move that Google was unwilling to make just two years ago. In the end, we don’t see anything in Pixel (yet) that justifies stepping on the toes of OEMs. Fortunately, the Pixel features an excellent camera thanks to Google’s killer software algorithms. The camera is fast, has killer low-light performance, and features video stabilization and 240 fps video. It’s what’s on the screen that counts.

The software on a smartphone is almost always more important than the hardware, though, and here, the software is excellent. This isn’t really new territory for Google; as with Nexus devices, you get Android from the creator of Android. That means this is the only version of Android that feels like a cohesive whole, with a consistent design and layout. The OS not only matches the unskinnable Google apps but also most of the third-party app ecosystem, much of which has adopted Google’s design guidelines. The software here feels like it is all cooperating to deliver a single vision instead of serving as a battlefield between Google and your hardware OEM’s software ambitions.

Android 1.0 was not an incredible piece of software, and Google Phone 1.0 is not an incredible piece of hardware. What made Android great was continual iteration and improvement over many years. Hopefully, Google’s hardware team keeps at it and one day justifies the decision to become a hardware OEM. For now, we say: welcome to the cutthroat Android hardware business, Google.

More at Ars Technica

Did you miss a roundup? Check the Eye On Open home page to get caught up with the latest news about open source and Linux.


Copyright © 2016 IDG Communications, Inc.

How to choose a low-code development platform