J5’s Blog

D-Bus version 0.62 was released. Thanks to Thiago Macieira of Trolltech for handling this release. Many fixes were added and we are now in the process of splitting out the bindings from the core. Once that is done we will be releasing a 0.90 API frozen core for your bug busting pleasure along with seperate binding packages which will not be included in the API freeze. 0.90 will be the last coordinated release between the core and the bindings. It is our hope to get a 1.0 core release out soon after that so that developers can start to rely on a long lived API/ABI stable D-Bus.

Release notes can be found on the main wiki page

The tarball can be found in its usual place

Luis, I have one point of contention to bring up about your concerns. We must remember what I think many forget. Companies, governments, organizations and individuals all make up the community. It is not the community vs. companies. What we see is that in any community (not just ours) there are good citizens and bad citizens and everything inbetween. Any entity in the community has equal potential to be pulled to either end of the spectrum.

That is not to say it is wrong to be weary of a companies motives. Because they do hold huge sway in certain aspects of the community, shifts in the way they interact within the community ripple more than say joe random developer. Of course there are individuals in the community who hold such respect that they also could cause similar effects. We need to protect against those who would be bad citizens and guide those to becoming good citizens. I think in the end the benifit of working together as a community will outweigh any singular interest as we grow the pie for GNOME.

Here are some of the notes I took from the LSB Summit. More to follow.

What is the LSB?

It is an organization dedicated to making Linux and open platforms more competitive as an ISV development platform. The idea to have a franchise modeled platform where the LSB provides the base standards and distros, upstream, etc. are franchise which implement these recommendations. The franchises then decide what goes into the next version of the LSB. The base standards give ISV’s comfort with working within the platform. Where appropriate ISV’s can go to individual distros for further certification outside the LSB’s scope.

Thoughts on what makes a good ISV platform.

* Broad Scope
* Predictability and Stability
  - Obtain ABI stability for Linux
* Developer Friendly
  - They have MSDN, we have Google - not nearly good enough
  - Newbies get scared

LSB Dates

LSB 3.1 (April 2006)
LSB 3.2 (Q2 2007)
LSB 4.0 (2008)

Issues

* GCC 4.1 libstdc++ v7
* glibc 2.4
* Language runtimes (Perl, Python, Java?)
* cross desktop (GNOME/KDE) interoperability
  - freedesktop.org
  - portland
* I18n (font management, input methods)
* Accessibility
  - can’t be bolted on
* Multimedia
  - Making it just work
* Packaging
  - Portable packaging?
  - Integrates with native packaging
* Printing
* Wine?

Other issues

* Improving testsuites
* Project infrastructure
* Franchised developer/certification programs
* Binary compatibility over time
* How does the LSB interface with all the key stakeholders?
  - Distros, projects, ISV’s

Some ISV musings

* compile once and run anywhere would be nice
* ISV’s can’t always control what their vendors are using (i.e. some inhouse fork of a distro)

Some Distro musings

* need better tests which work on all their supported platforms
* don’t want standards which overly complicate the platform

Some Upstream musings

* doesn’t want standard pushed on them
* wants to figure out how to work with ISVs

It is late so I’m just going to provide a brief commentary. I most likely won’t get to my notes until this weekend.

First I would like to give a shout out to Aaron Seigo of KDE fame. We hung out today and I can say he is the type of person we need more of in the community. He is energetic, extroverted and shows a passion for life and the community. I’ll be going to aKademy this year if only because he owes me a beer.

I went to the LSB Summit as a GNOME Representative. Here is my immediate reaction to what I heard there.

Things I liked about the LSB:

• desire to support ISV’s without cannibalizing the community - standards where they make sense as opposed to standards for standards sake
• open dialog on the issues - I learned a lot including ISV concerns and accessibility issues
• The ability to be candid with a KDE counterpart and find common ground

Things that scared me at the LSB:

• Trying to please too many people - Multimedia and printing begin to sound like - If we can’t decide on just one lets support them all. Pick one or pick none and wait until one becomes the clear winner.
• Portland Print Dialog - Perhaps I don’t understand it since I missed the first printing summit but UI by committee usually ends up in disaster. Where do we stop? Portland file chooser? Portland widget toolkit? Let technologies show up in Portland first then I think it is appropriate to discuss them with the LSB, not the other way around

[read this post in: ar de es fr it ja ko pt ru zh-CN ]