Announcement: Genode OS Framework version 14.02 released

Norman Feske norman.feske at ...1...
Sun Mar 2 13:13:46 CET 2014

Hi Lonnie,

I agree on the points you made for making Genode more popular. The live
system will certainly raise much more awareness of our project than the
regular releases.

However, I see popularity as a two-sided sword. Of course, getting more
attention increases the chance for more people joining our community and
bringing forward the technology together with us, which is ultimately a
good thing. Maybe it will even foster the consideration of Genode for
products of companies that are unaware of Genode yet.

On the other hand, I am afraid of two things. First, the live system has
to be really good and convincing to developers and end users alike.
Otherwise, people will give it a try, it may leave a half-baked
impression on them, and this impression will stick for a long time. Some
people are quick with their judgments, in particular end users. In
contrast people who spend the effort to download and compile source code
are less likely to draw conclusions too quickly. The latter I would call
our target "demographic", which I feel safe to address right now. The
live system widens the audience. So we have to address the topics
discussed in issue #1000. And we need to pay attention to thorough
testing and little details, which is a lot of work in addition to the
actual features to present.

The second point that worries me is related to Daniel's email. Once, the
live system achieves to get people hooked on Genode, they may have high
expectations on how easy it is to customize the live system. If such
people download the source code but then hit a wall, this experience
will leave a bad aftertaste as well. To see what I mean:

Such a bad experience can do our project more harm than it helps us,
especially when it becomes an universal truth through public postings
like the first comment of the thread above. So we either have to clearly
communicate that the live system is just an example to show off Genode's
capabilities, or we have to make it really easily "hackable". I would
prefer the latter. But making Genode easier to use is hard. In
particular because "easy to use" for us developers has a different
meaning to "easy to use" for, let's say, a GNU/Linux power user. The
current "make prepare" and "make run/..." mechanisms are a good start
but still far from being easy.

Please don't get me wrong. I don't want to shy away from the live
system. But I regard it as a very sensitive topic. I certainly want to
make Genode more popular, but we have to make sure to leave a
sustainable good impression, not just on our immediate target
demographic but on a broader audience. So we should give it the time
needed to make us developers feel confident with it.

Best regards

Dr.-Ing. Norman Feske
Genode Labs ·

Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden
Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth

More information about the users mailing list