Mail Archives: geda-user/2019/02/13/03:21:45
Hello Bert,
On Wed, 13 Feb 2019, Bert Timmerman (bert DOT timmerman AT xs4all DOT nl) [via geda-user AT delorie DOT com] wrote:
Thanks for all the good wishes, the same to you!
I am trying to focus on the specific statemens/questions below.
> I have learned from that period that pcb users are using and not developing,
> sometimes contributing a (large) patch or bug report, which are most welcome.
The pcb-rnd community is a bit different, probably more active: we have a
core power-use plus developers group who generate a lot of relevant
traffic on daily basis.
This is mostly happening on IRC, so not visible in svn or mailing list
archives.
>
> I also learned that the decision for a fork was made before it was even
> discussed with me.
The pcb-rnd fork decision was made back in 2013. I was active on the
mailing list since about 2005 and tried to join coding too. I was
maintaining pcb-gpmi (the scripting upgrade) since about 2010. I don't
think the fork was a surprise for anybody back then.
(IIRC you did not lead the pcb project at that time. I didn't have a time
machine so I didn't know you would, that's how we didn't discuss it before
the fork.)
Anyway, it doesn't depend on one specific moment (e.g. moment of fork
decision): there were multiple chances to align pcb-rnd and various
subprojects of gEDA (including a potential pcb merger back in 2016), but
it never really worked out. In the same time it was super easy to build
working cooperation with some non-gEDA projects, so I guess it's just some
fundamental incompatibility. Maybe both in where an EDA suite/ecosystem
should arrive at long term and what path is leading there.
So it's better to move on.
> BTW: I noticed an enormous amount of commits have been done in pcb-rnd over the
> last 3 .. 4 years, tens of thousands of them, my compliments to you. I guess
> you may have retired from your day job to be able to do that ?
Not really. I wish I could retire, but that's many many years from now!
Instead I have an almost normal daytime job, in a conventional small
company office and 2*40 minutes commute a day.
The tricks are that I spend most of my free time on pcb-rnd and we have a
real active core team who do a big chunk of the work and we have a system
where administration costs are low.
About the workshare: very often for a complex bugfix you will see 2..4
commits I make, but that's like 5% of the work - the 95% of the work was
done by a fellow pcb-rnd developer or user who figured what exactly
triggers the bug and crafted perfect minimal test case.
Regards,
Igor2
- Raw text -