Recently DesktopLinux published it’s 2004 Desktop Linux Market survey which confirms my feelings about current direction of Linux market.
First is overwhelming domination of Mozilla based products in the area where they exists. Thunderbird and Mozilla Mail have ~37% early beating Evolution and KMail. Firefox and Mozilla holds over 70% of browser market. And, both Thunderbird and Firefox were released in Q4 2004! We can easily predict that this numbers will grow in 2005. Some of my thoughts:
Results shows the growing usage of SuSe distro which came second. In my opinion SuSe became the easy distro. It’s extremely usable, easy to install, full of ready_to_use features that are easily adopted by new users. I think that SuSe is the distro which should be offered to every newcomer as an example of what Linux can be.
Another thing is that big-3 – RedHat, Mandriva and SuSe holds ~50% of market and first three positions. All of them are company-driven projects. It shows that Linux is not a bad place for making money, and creating open software can bring you money. It’s also a prove that what Linux really needs is a usability part. Those distros are extremely focused on usability, on transparent adoption and migration from other platforms, not made for geeks, nerds and folks like that but for end users. Nice adoption of Slackware and it’s younger brother Gentoo shows polarization on the Linux market. You can focus on end users, geeks or you can try to be a base-platforms for other distros.
You can also split them into group of company-driven products, and free-of-core ones. SuSe, RedHat, Mandriva vs. Debian, Gentoo, Slackware. I find it extremely good to have such strong roots. Every of those 6 gives a user something different, and every user can choose. I think that all off them will live long and well.
Browser, Email, Office
I think that most of you will agree that those three kind of apps are the very most important part of your desktop. I think that for most ordinary users those 3 are used for 95% of time spend in front of monitor. There are others like IM clients, music clients and so on, but their API is less important as they are used in very simple way (with no offense to their authors). Usability of web browser is crucial. Imagine how much time you’d lost if you would have to make one additional click to open every new webpage. That’s why I believe that Firefox is such a great success. Because it’s not only easy to use yet powerfully, but also solving UI problems is relatively easy and I think that all bugs major UI bugs reported about Firefox 1.0 will be solved by this year! I see a potential of Firefox mostly because of it’s UI. Behind Firefox stays Gecko – Gecko is even more great piece of software. And I’m excited about my imaginations about the future use of Gecko with cooperation with Gnome and KDE guys. What I miss in that survey is info about office usage. I’m wondering how many users use Microsoft Office on some Wine-like, how many use OpenOffice and how many use KOffice. Anyone has some results?
KDE, Gnome and others
Window Environments are walking in the same direction as the rest of Linux software. Or, we can say that they’re pointing this direction. With a huge dominance of KDE&Gnome we have a power of choice, we have two stable and actively developed GUI libraries, we have competition and we have cooperation in the name of users comfort. We have product that bases on company-driven library, and totally community-driven project. Two models, two ways of development, both with their advantages and both the way of Freedom.
KDE wins not only because two main distros use it. It winds because it’s more candy and, whatever you say, users like candies. I’m a KDE users and I love it’s style, if I would have to advice some WE for Windows/MacOSX user, it would be KDE. KDE will be most used Linux WE for a long years, and Gnome will be for those who doesn’t like it.
So. After looking at those stats I see some things we (as a Linux users/coders community) could/should improve. Users chose KDE and users chose Firefox. In my very humble opinion it should be a sign for people behind both projects. Sorry, but KDE does not win by 2% over Gnome, but by 40%!!! And I’m sure that KDE and Firefox will gain even more market in 2005. Even blind would see the problem. KDE must work well with Firefox and Firefox must work well with KDE. And there are two reasons this doesn’t happen.
The first one is that most Mozilla coders use Gnome and know GTK. It’s OK, you can’t force someone to code something in our world. So we need KDE/QT coders in Mozilla project. We need people who can not only port Firefox to QT, but who can keep an eye on usability issues so KDE users will feel good using Firefox. Second reason is that KDE has it’s own KHTML engine, so some KDE coders might feel that it could lead to clash.
And I have to agree. Unfortunately I think that areas of two KDE projects overlaps with some standalone, bigger projects with greater potential. It’s KHTML and KOffice. KHTML is smaller, faster and probably cleaner than Gecko. But Gecko is fixing them with a nice speed. And looking in the future I see no reason for KHTML existence. Gecko does it all, and does much more. And with help from coders who’re working on KHTML it could reach this level sooner. Who could benefit? Users. Both Gecko and KHTML support standards almost in 100%. So there is/will be no difference in user experience in term of viewing web pages. User like to have a choice, so he have – Konqueror, Firefox, others…
Now, when we know that KHTML does not benefit much from Apple’s Safari I feel pity thinking of the small group of people working to code something that is already coded in Gecko. And what will happen with CSS3? Xforms? VoiceML? SMIL? Other standards? They will have to code it by themselves? In this example, I believe that it would be better to have them working on Gecko. KDE would benefit, users and Gecko. Zack Rusin is planning to port Firefox to QT4, and I believe that this will be the moment for KDE to move from KHTML to Gecko.
Second area is KOffice. It’s nice, usable, and have some advantages over OpenOffice. But OpenOffice is improving every day, and as a KDE user I’d prefer to see OpenOffice integrated better with KDE.
This year shouldn’t change much in those stats. FWIK there won’t be any new players, KDE improved with 3.4 and will improve a lot with 4.0, Firefox will improve with 1.1 and 1.5, SuSe will release probably 10.0 by the end of the year, and it’s 9.3 has great press… Improvements, not revolution. I still predict that Linux adoption on end user market will begin in 2007. So we have time to get ready…