Improving The Linux Desktop: 20 Needed Fixes

Monday Nov 1st 2010 by Matt Hartley

The Linux desktop needs to focus on new users, improve wi-fi tools, work better with commercial software, among other necessary changes.

I love using Linux. I enjoy the control over the computing experience it affords me and how I can choose exactly how my desktop is to be run.

I'm not some guy writing Linux articles on a Windows box or a Mac. I "live Linux" exclusively, every-single-day. And I have done this for years.

That said, I’ve put myself into the position of those who are new to desktop Linux or simply overwhelmed by something within the platform. With this article, I hope to address new ideas alongside some challenges that I believe, if dealt with realistically, would make using Linux more accessible for everyone.

1) Focus on the casual user first, geek second. Many among you will likely point out that nine times out of ten, a Google search holds the answer to the most common questions people have when working with desktop Linux.

Unfortunately, not everyone out there knows the right questions to ask Google in the first place. What’s needed is some kind of easy-to-use GUI troubleshooting tool that can be used to gather debugging information. This would make a trip to the various Linux forums a lot more productive for everyone involved.

2) OTA data sync options. I think we're going to see more companies (than just Google and Canonical) heading in the direction of over-the-air data (OTA) synchronizing options.

Despite the fact that Google has managed to take some of the sting out of syncing our smartphones with our Linux desktops, the sad fact is that users of the iPhone are often an iOS release behind when it comes to keeping our music in sync.

The only viable solution I've found is the new service offered by Canonical with their Ubuntu One music hosting. The service allows me to listen to my tunes without needing to sync anything locally, instead doing so remotely from just about anywhere I choose.

The problem is the software provided for the iPhone is very young in its development and lacks critical features like continuous play and playlist management. Now imagine if more resources were dumped into Canonical's efforts with Ubuntu One. The possibilities are endless, once we’ve dealt with the issues seen currently today with Ubuntu One's music services on mobile devices.

3) Less software politics. Opinions – everyone has at least a few passionately held ones. And the passionate opinions surrounding FoSS vs. proprietary software tend to fuel essential debates about this topic.

Where I see a problem is when this raucous debate causes casual users to balk at adopting Linux on the desktop. See, when you ask most people off the street, they couldn't care less about the politics of software. What they're interested in seeing is how a community can work together to provide support for one another and how usable Linux has become for most people looking to try something new.

So while it's good to debate the merits of FoSS over proprietary code, I welcome the day when we can see less protesting and complaining and more focus on demonstrating – with action – how FoSS is actually better.

4) A rebirth of specialized distros. Prior to Ubuntu's big success, it seemed like there were a lot more specialized Linux distros available. Everything from distributions formulated for security needs to distributions designed for network attached storage (NAS) appliances.

While some still exist today, we never hear about their merits due to a lack of media coverage. Everything is about the one single distro to meet everyone's needs. It's rather unfortunate. I would love to see a rebirth and refocus taking place on specialized distros. Imagine using these to target sales efforts to enterprise audiences such as radio stations, TV networks and other situations where having a specialized software toolset is desired. We did it in security and network storage, let's do the same thing for more niche areas of business as well.

5) More media focus on other non-Ubuntu distros. As you might imagine, not focusing so much on Ubuntu is difficult to do when writing about Linux on the desktop these days. Regardless, there's a clear need for us to spend more time searching around the Linux eco-system to rediscover other Linux options we might have forgotten about.

Why is this important? Putting all our eggs into one basket is working against what Linux is all about – freedom. I think a singular distribution of Linux would be a threat to this freedom. And with a possible loss in Linux freedom comes users having a poor experience when Ubuntu doesn't work as they expect. This is an issue we must all face. It’s already showing signs of hurting Linux as a viable operating system.

6) Make things just work. Test a Linux release before sending it out to the world at large. Unlike other distributions that take more time in the testing department, lately Ubuntu has let bugs slip that were, honestly, ridiculous.

The recent Intel "tiling bug" comes to mind. I would love to see less development speed from Ubuntu and more focus on getting it right the first time by taking things a little slower. While others will merely point out that there is always X distro instead, the fact is that Ubuntu has built up a brand that represents Linux collectively these days. So how about not making the rest of the distributions look like fools, fair enough?

And if the answer is to simply yield to using long term support releases only, then how about a website that reflects this?

7) A compatibility checker CD. Sorry, the LiveCD isn't cutting it. Not everyone thinks to check every device attached to their computer before installing the operating system. And if there are drivers available for a problem device outside of the included software repositories, knowing this ahead of time would be awesome.

Imagine a CD I can drop in that acts as a hardware compatibility checker! It would check the databases from bug reports about what’s working using tools like CUPS and SANE. I refuse to believe this is that big of a stretch. It’s been available for years on other desktop OSes.

8) Mall kiosks and trade show events. I've never been a huge fan of the "county fair scene." Same old booths, same old livestock and same, tired-looking vendors selling junk no one really wants or needs in the first place.

Now imagine happening across the same event with a booth showing off MythTV or even better, Fedora on a new laptop using mobile broadband! VGA connect that computer to a big screen TV and, I promise you, people will be doing double-takes left and right. Questions will come up and inquiries will begin to build before you know it.

Now imagine that this is done by the local Linux User Groups (LUGs) instead of investing 100% of their club resource into Linux Festivals. As nice as "LinuxFests" may be, you’ll likely gain more ground faster, expanding the reach of the Linux community farther, by leaving that comfort zone and hitting the streets with real desktop Linux demos.

9) Finding a way to monetize. I hate to say it, but the "donate now" buttons offer a poor presentation. If it's a hobby software project where these buttons show up, fine. But for those who really need the money to compensate them for their development efforts, something else needs to be explored instead.

Software as service (SaaS) is still a viable option for many open source projects. Other projects would do very well with up-to-date how-to guides for sale. The problem with the latter however, is the issue of software updates and changes. Then again, this also presents a fantastic opportunity for a subscription to assistance in a paid format.

However it's implemented, the key factor is that we need to find a way to help open source developers pay their bills while doing what they love. A PayPal donate button is simply not a sustainable solution.

10) Documentation. Some Linux distributions are better than others in this area. And perhaps the solution falls back to number #9 above.

Yet whatever the fix is, we need to find a way to get documentation on track for today's Linux distributions. As it stands, the quality of documentation for select popular distros (which will remain nameless) is a sad state of affairs. Some simply need to be set ablaze and rewritten.

11) A serious approach to wi-fi. We're creeping up on 2011 and people are still fighting with Linux wireless. The really funny part: this is completely avoidable.

The fact is, not every internal wireless chipset is going to work easily. Yes, sorry to break it to those fans of NDISWrapper, but there are some wireless devices that just need to be replaced. The solution I've been begging for is for Linux enthusiasts to stop supporting those who don't support us. Hit the wireless vendors in the pocket book!

In the meantime, there are indeed USB chipsets for 802.11n that work great, from both the Ralink and Atheros product lines. I've spoken with Ralink devices vendors such as Edimax in the past and they're not against the idea of a branded Linux dongle. They simply need someone of means to step up and get this rolling. I use a number of their dongles. They all work pretty darn well.

12) Embracing Windows repair techs. Contrary to popular belief in FoSS circles, repair techs who use Windows are not evil incarnate. Most Windows techs are simply trying to earn a living doing what they love. For us to not embrace them and try to get them to see the value of adding Linux to their list of offerings for potential clients is a really big missed opportunity.

When these individuals see their client for the eighth time in a month to remove the same malware all over again, wouldn’t they look better if they offered up Linux as an alternative? While it might appear to lessen their earnings at first, I bet they could make up for it in new referrals.

13) Dedicated home partitions. For experienced users, nothing could be easier or be more valuable than a dedicated home partition wizard as a part of the initial Linux install. A default option to create a suggested home partition size based on available hard drive space would be a huge selling point for any OS.

Surely this isn't too much to ask? And yes, dedicated home partitions mean less data loss as an added bonus.

14) Roll back to previous distro release. With the possible exception of creating dedicated partitions for the home directory out of the box, the other most needed missing feature is a "roll-back to previous distro release" option.

I cannot even begin to express just how badly needed this option is. I mean, bundle this with the dedicated home partition, and any user would feel bullet proof with their Linux box. Not having this as a default option limits the power of the Linux desktop compared with other alternatives.

15) Play nice with proprietary software. The realization that proprietary and open source will have to play nicely is quickly become an uncomfortable reality for some folks. My take? So long as it's not installed by default, who cares?

As long as the user must install it themselves, it's not bothering anyone. Complaining every time Ubuntu or another distro comes closer to embracing the proprietary software world doesn't fix anything. And it sure hasn't stopped folks from asking for Photoshop for Linux, among other applications.

16) Browser choices out of the box. Offering different default browsers by choice during a Linux install is a great next step. Obviously it would make sense to provide the advanced user install options to bypass this kind of thing. But a full menu would provide new users with the immediate sense of freedom in seeing their browser choices right there.

This is not a critical need, rather something I’d like to see. It would make folks feel more at ease if they were using Chrome in Windows and didn't know where to find it for Linux otherwise.

17) Microsoft Exchange alternatives. There are countless open source groupware suites out there and it feels like no one is using any of them in the enterprise situations we work in.

Getting Zimbra or something like it ready to go with simple OTA solutions would be one option. At the end of the day, too many of us are waiting on Google as a possible alternative and it's not even close.

18) Real marketing talent. You know, I've seen the "try Linux videos" created for various contests over the course of 2010. In each instance, the exclusive focus has been all on the "freedom" aspect and the producer failed to realize the users’ true needs.

They’re not speaking to Joe and Jane Average. How about focusing on the time savings in quick Linux installations, freedom from driver CDs, or being much less of a target for malware?

19) Long live LibreOffice! Let's find a way to push some real money into the LibreOffice project. With the people at Oracle out of the picture now, how about if we seize this opportunity to give this localized office suite a massive design make over?

While the option to deal with formatting in DocX snafus is beyond our control, the fact that we still have something that looks like it's from 15 years ago is just pathetic. Let's incorporate some new graphics and maybe even….some color? Open Office already has a new name with LibreOffice. A new look to go with it is the next best step.

20) Just be Linux. Distros need to focus on the common sense stuff. Let’s have less focus on being like OS X or Windows. Focus on what is already working and stop trying to reinvent the way it looks to make it "prettier."

While some limited visual updates are welcome, the desktop managers need to stop thinking like coders (exclusively) and actually look at the usability of what is being offered. Instead of three clicks, can this be done in less? And for the love of decency, Ubuntu, no, Unity isn't what I had in mind.

Mobile Site | Full Site
Copyright 2017 © QuinStreet Inc. All Rights Reserved