Roadmap 2019

[John J. Karcher]

On 12/23/18 2:46 PM, Norman Feske wrote:
> Hallo everyone,
> 
> hereby, I'd like to follow up on our tradition to discuss Genode's road
> map on the mailing list. Let's take the turn of the year to recapture
> the events of 2018 and make plans for the next twelve months. Please
> feel welcome to chime in! By mid of January, I'll finalize Genode's
> official road map and would like to take your input into account.

[snip]

> Please don't hesitate to share your thoughts. You may consider the
> following questions:
> 
> * What are your ambitions for 2018?

I have goals in three basic categories:

1. Desktop.  This has two parts.  The easy one is to switch to Genode as 
my VirtualBox host.  But I also would like to have the VirtualBox Guest 
Additions for using Genode within a VM.

Would this qualify for the roadmap?

If not, I am starting to work on it - it should make a great learning 
experience (and hopefully a useful blog series), because the various 
facets touch so many different parts of the system.  But I will be slow, 
so if others are interested, I will be happy to learn from them instead! 
  ;^)

2. Network Appliance.  I would like to run an always-on system (probably 
a Raspberry Pi or something similar) for a small mail server (dovecot, 
etc.), a file server, and a streaming audio server.  This should mainly 
be a matter of porting the appropriate software.

In order to run this machine headless, I am (slowly) porting/writing a 
VNC server.

3. Tablet.  This is part of a longer-term goal, but I would love to have 
Genode running in some form on an open-hardware tablet if possible.  (I 
don't even care if it's useful at this stage.)

I know there are efforts toward building mobile front-ends under Qt, but 
I would be even more interested in working toward a native UI front-end. 
  I really feel that Genode would be a great fit for tablets.

I know tablet-related work will not make the roadmap for 2019, but would 
it be possible to agree on a semi-official tablet hardware target for 
the next year or two, for those of us who are interested in pursuing this?


> * Which areas of Genode would you like to see improved?
>    How would you possibly contribute to these improvements?

- GUI system configuration (e.g., adding configuration features to the 
Component Graph, better discoverability for Depot packages), as you 
already described

- I would feel more comfortable with my VM host running on a journaling 
FS (e.g., ext3 / ext4).  Do others feel the same way?

> * Which technical topics do you find tempting?

Almost everything.  (This works both for me and against me!)  Despite 
having mentioned ported software several times, my main interests are in 
exploring the native Genode APIs, proper component design, and 
verifiable code techniques (e.g., Ada/SPARK, etc.).  I'd also like to 
start experimenting with RISC-V.

> * If you imagine a Genode-based system one year in the future,
>    how would it look like?

Good question.  I would see it very similar to today's, but with more 
applications (ported and native), along with the beefed-up GUI 
configuration tools mentioned earlier.

> * Do you have further ideas that would help making Genode relevant
>    at a larger scale than today?

Sorry for the boring answer, but I agree that there aren't really any 
technical barriers any more, so I think your ideas on highlighting 
Genode-based products/solutions and Genodians.org are the right direction.


  Thanks!

   John J. Karcher
   devuser at alternateapproach.com