I was present at the 2004 GUADEC (GNOME Users and Developers Europen Conference) in order to represent Debian on the GNOME advisory board (at the request of the DPL) and to talk to companies active in the Linux desktop community. Several things were brought up several times by different people:
1) Many people use or have used Debian in the past and respect Debian to a surprising extent. There are some disagreements with respect to individual DFSG judgements, but the general principal of aiming for an entirely free distribution is seen as a good thing. 2) People believe that our release cycle is (a) too long and (b) not predictable enough. Part of this is due to the contrast with GNOME's 6-monthly time-based releases. 3) The way the DFSG is currently interpreted by debian-legal is not obvious to an outsider, and some interpretations are felt to be excessively extreme. Some companies feel that various licenses were genuine efforts to be DFSG free, but the discussion that followed their release was sufficiently confrontational to reduce any desire to fix any bugs. It's been suggested that the combination of 2 and 3 has led to us missing several opportunities for wider deployment, and they're probably issues we should look at in the future. More positively, the feeling is that the quality of our packaging is good. We're actively engaged with upstream and doing a fairly good job of pushing significant patches back, and there's a general perception of us as good community players. This is largely thanks to the efforts of the Debian GNOME team, who have done a great job of improving our relations. I raised two main issues at the meeting itself: 1) The use of copyright law in an attempt to protect trademarks. This is potentially going to be an issue for us, as it leads to artwork that we can't distribute in main. This is also less than ideal for upstream projects, as it reduces the level of branding and general market awareness. However, if we want to make any significant argument here, it seems likely that we need the ability to make a solid legal argument as to why copyright law is the wrong way of handling this. 2) The possibility that GNOME's adoptation of Java as an application development language would result in software depending on a closed JDK.The consensus appears to be that GNOME will never ship code that can't be run with free Java implementations. As things currently stand, it looks reasonably likely that we will be the first major distribution to ship with Mono 1.0 and surrounding libraries. This gives us an opportunity to be seen as a good development environment for the GNOME platform - however, to be able to take advantage of this, we are likely to have to reduce time between releases and make our approach to licenses clearer. -- Matthew Garrett | [EMAIL PROTECTED]