This is a great blog post by +Lennart Poettering
. In fact, I put in a talk at +The Linux Foundation
Linuxcon Europe that this blog references. However, my talk talks about the viability of a the Desktop as a business market. What is missing is that we have no analytics about the strength of the market as it pertains to the desktop. The 1-4% desktop share that we purport to have doesn't really have enough data.
If you are running GNOME on an enterprise setup, using RHEL or SLED, the +GNOME
has no way of knowing who uses their software. Application writers have no personal relationship with the desktop mraket because it is completely defused through the distribution network. Quality issues also arise. One distro can have a bug that is no repeatable in another. IfI have a problem, do I blame the distro or upstream?
Centralizing application distribution, with sandbox'ed app, provides desktops like +GNOME
and application writer with the kind of analytics that will help improve software quality and reliability on a wide range of platforms. If an app fails we know exactly who to blame.
In many ways, Lennart's talk and mine (if accepted!) would be complimentary. Of course this might mean that Lennart either follows me or I follow him, and boy I know how that ended up, last time that happened! :-)
I think it would be awesome to have a conference where we place these kind of things as the main focus. It seems that a lot of conferences these days center around kernel development.