Drive:Activated logo
hi there!

I see you've stumbled on to my humble home on the net, Drive:Activated. My name's Sam, I'm an ambitious and driven uni student, residing in Melbourne, Australia, wanting to make my mark on our world. This is my site, which is mainly just my blog and some other bits. There's no definite theme to my blog, just anything that interests me, and currently that's web trends, startups, ideas and cool stuff. Check it out, leave me a comment, click on 'Who is this?' to find out more about me, or drop me a line by clicking on 'Let's Talk'. Hope you enjoy it!

My signature

Content sign

Fueling the Linux battle practically - part 3

   Filed under: ,    
This is the last post in a series of posts about making Linux a more competitive in the consumer world. The first post, part 1, introduces the reason and some of the problems and solutions, like lack of innovation and boring UI design. Part 2 explores more solutions, including eliminating the restricted formats issue, improving the out-of-the-box experience and connecting the web browser to repository apps. They're not requisite reading, but worth reading I reckon (because I wrote them Stick out tongue).
  • Compilation is a no-no. I should never have to compile anything that's released to the public (unless I'm doing something that absolutely requires it, like coding specific custom functionality, but then this is moving in to the geek domain). They should all be as easy to install as downloading them, launching them, and answering some questions if really necessary.

    This problem occurs mainly because when developers release apps, they don't have the time or resources to compile it for all the popular distributions out there - ultimately, there'll be people who'll miss out and the only way to get the software is to compile it themselves. That's why the recently publicised plan to simplify software installation across all supporting distributions by the Linux Standard Base project is such a great idea. "[ISVs want] to treat Linux as a single platform, which means they want to offer a single package for Linux, much as they do for Windows."

    There is also a time lag between the developer releasing a new version and the distribution's repository having an updated version - a time lag that sometimes people can't wait for (and for which there is no known timeframe for), hence forcing them to try compiling it. This is a disadvantage with the repository style of distribution applications that can be reduced if the developers ran their own repository source and the above LSB project succeeds - devs will then only have to do one release and all supporting distributions will have the newest release.
  • Create alias for the file system root folders. What does /etc, /dev, /var, /bin, /root etc. mean? If I wanted to install a program manually (because the program doesn't come with an installer) where do I install it to? They should only have to deal with the home directory ideally, but these people are hobbyists, tinkerers - they will look around and the more you let them know what they're doing, the better. I understand it's a Unix legacy, but a better solution would be to create alias for each of those folders and hide those folders in the GUI. For example, instead of /home, they could call it /Users. Instead of /etc, they could call it /Configuration Files.

    Linux file system
  • Improve interoperability with non-Linux computers. In both Gnome and KDE in OpenSuSE 10.2, the shell smb:// support is still non-existent in most apps, e.g. Banshee, so the only ways to access files on another computer is to copy it to a local folder, or use mount in the terminal. Of course, Nautilus and Konqueror, Gnome and KDE's file manager support this.
  • Reduce or hide choice. Geeks will scream at me for saying this, but when even the basics can vary so much, it becomes,

    a) very difficult and time-consuming to work out what I want (is Ubuntu better than Fedora Core, or is Gnome better than KDE?),
    b) very difficult to trouble shoot, especially over the phone, and
    c) confuses people when a certain program is not available for their distribution. The nature of the Linux ecosystem makes this very difficult but without any sort of across-the-board standard, trouble-shooting becomes very frustrating (e.g. only some distributions have pump & sax2, others have equivalents), and instead of knowing how to work Linux, you only know how to work a specific variant of Linux, reducing the community's ability to help everyone.

    Some sort of standard that dictates what command-line utilities must be included would be good too across all the main distributions. You can put more in obviously, but those basic ones must be there, so for example, there'll be no need to go digging around to work out what the command to renew a DHCP IP address is for every distribution.

    Also, distributions should be made such that you can change distributions without having to reinstall everything. If you give hobbyists the choice, they'd want to be able to switch. But if you force them to reinstall everything every time they want to switch distributions, they're not going to be happy.
  • Stop labelling applications as for KDE or for Gnome if they can run on either (which most can). That labelling just confuses people, who are already struggling to grasp what KDE and Gnome are anyway as there is no equivalent in Windows or Mac (unless you count the terms Aero and Aqua respectively, but they're hardly used).
  • Make the Linux Kernel version prominent. This will confuse users, and therefore some explanation is in order, however it is necessary because the only constant that hardware manufacturers can rely on is that for their hardware's system requirements, unless they're willing to test their product on all the popular distributions out there and list all of them.
  • Finally, don't aim to make it as good as Windows, Mac OSX or anything else out there. That's a milestone. Make it better, unique and easy to use in every way and you'll have a winner on your hands (or in everyone's hands).
So Linux people, stop wasting your time spreading FUD (and distance yourselves from the communist fanatics at FSF who can't sell something even if their lives depended on it) and start making Linux a viable alternative to Windows. Its not a war, its competition - use your competitor's resources to leapfrog them if you can instead of reinventing the wheel.

For those who can't seem to be able to shake off the FUD spreading and the microsoft-bashing attitude, take this from someone within your community (its in relation to the recent Novell/Microsoft deal, but its applicable in general too) -
"I understand that many people don't like it as Novell is collaborating with the 'evil empire'. But I don't like this way of thinking. We are not working against somebody, but we are working for Linux. Fundamentalism always leads to pain. I think it is a good thing, especially for the users."
-Hubert Mantel, co-founder of SuSE Linux (quote from CNET, emphasis theirs)
Trackbacks sign
No Trackbacks
Trackback URL
No trackbacks yet - link to me people!
Comments sign
No Comments
Comments RSS RSS icon
Come on, be the first to unleash those thoughts from within.
Post comment sign
Leave a Comment
I know you want to!
(required)  
(optional)
(required)  

Want to keep stay in the loop with the comments here? Leave your email address below and you'll be informed when a new comment is added to this blog post.

(optional):  

Submit