On Fri, 2009-05-15 at 08:26 +0200, Øyvind Harboe wrote:
> My interest lies in a couple of things. I don't know what
> "role" this translates into though:
> 
> - recruit first rate maintainers. Actually I'm kinda done with this as
> it seems we have a positive feedback loop now.

This is "Recruiter", and you can never be done with that job.

> - commit good patches to svn quickly so as to encourage contributors

I am seeing the need for at least one "always open to commit" testing
branch that allows this kind of behavior, but this is off-topic.
These topics will get their own thread soon enough.

> - keep an eye on performance. OpenOCD has a unique capability
> of supporting *both* high processing power high latency(USB & TCP/IP)
> interfaces *and* low processing power low latency(embedded hosts).
> The policy of OpenOCD is that all interfaces are equal, but if an
> interface is no longer maintained or tested, then the community
> has no responsibility for it. I like that: only if we take responsibility
> do we have responsiblity :-)
> - there are, still, architectural problems with OpenOCD.
> I'd like to make these changes if they make the code more clear,
> reduces # of lines of code, more robust to changes(i.e. that changing
> X doesn't break Y) and there is a reasonable way to transition
> and retest.
> - target testing & qualification. I want a list of supported and tested
> targets/flash devices. It is not so much that I want everything to
> work than that I want to know *what* I can claim to work. It's OK
> if an older version has to be used for some obsolete target
> hardware.
> - I want to get the basic architecture and inner workings of
> OpenOCD right & stable so we can have long term stability
> in place(years). I believe that a couple of years from now we will
> recall the good old days of '09 when we had virtually no
> users, few targets and we could do what we wanted. Now
> is the time when  we can make changes relatively easily.

So to sum these up:
- performance
- architecture
- testing/qualification
- stability

I think all of these goals fall under "major contributing developer", if
not "systems architect".  I think that everyone needs to share this
later hat in a meritocratic fashion: in open source, those who propose,
seek consensus on, and implement major changes are the Architects.

However, I think there is a more important role that you play, which
needs to be called out -- if only to draw out others from the same bush.
Given that Zylin AS produces the ZY1000, you are an OpenOCD "Vendor".
Right now, I am not sure what that means, but it deserves mention.
In fact, I think this point deserves a thread of its own.  Stay tuned.

Personally, I think that Zylin AS has been an excellent OpenOCD citizen.
Providing support for the community is frequently a thankless task, and
you have done a first-rate job of communicating with and supporting the
community.  Despite any mistakes you may have made recently, I think you
deserve high praise for your long-term commitment to the project.

Cheers,

Zach

_______________________________________________
Openocd-development mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/openocd-development

Reply via email to