kicad-developers team mailing list archive

[Rick Walker <walker@xxxxxxxxxxxxxx>]

Hi Dick,

> I guess if I was on commission I would try and talk you out of your
> decision. 

I didn't make any decision.  I still use kicad for my own consulting
work.  I just couldn't get the team at Corning to stick with it.  I
tried to talk them out of the decision. 

I'm hopeful someday to reprise the decision and maybe get a
different outcome.  Kicad seems to be making great strides.

> I am honoured that you at least think I had something to do with KiCad.

I guess I mistakenly got that idea from several places:

Firstly just looking at my nmh mail log going back to January 2011:

   scan +kicad | awk '{print $4}' | sort | uniq -c | sort -n | tail -10
    113 Adam
    120 Fabrizio
    144 Cirilo
    197 Brian
    215 jp
    252 Miguel
    346 Wayne
    556 Lorenzo
    897 Dick

Looks like you almost trump the next two authors combined in total
postings. 

Secondly, your name is second only to Jean-Pierre in the Developer's
credits under Help->About_Kicad.  Any idle passer-by can immediately see
that you are a shaker and mover here. 

You're certainly the most vocal and opinionated on this group (I've been
following the discussion for several years) and generally speak in
absolutes regarding your vision of the future of kicad.  If it's good
enough for you then the rest of the world should fall in line.  Of all
the dogs, you definitely bark the loudest. 

> I don't know how I'd feel if I ever ran into a bug in Allegro. 
> Probably pretty damn annoyed, and helpless.  It's that fear of
> helpless-ness that keeps me in KiCad. 

Generally, Allegro ships a bug-fixed stable version that has undergone
pretty thorough regression testing.  The Corning team has found a few
oddities, but nothing show stopping.  Allegro also has paid support,
being a commercial product.  Just the fact that most users are using a
"stable" build makes the bug problem much less likely.  They are quick
and prompt with back-patches when real issues arise.  Most of our team
feel much more helpless with an open-source product, unfortunately. 

Not everyone has enough neurons to design a 10Gb/s fiber optic
transceiver and protocol conversion system in a 5x25mm footprint and
still have bandwidth left to track down errors in Kicad's via handling
code.  If it isn't stable enough to do a design at 95% confidence level,
then it isn't stable enough to use at all. 

> If I was having problems with Fedora, I'd dump it.  Generally I like
> to identify what is the tail and what is the dog.  Generally the dog
> should wag the tail and not the other way around. 

That's an incredibly kicad-centric view of the world.  Let's say that
Fedora is 100% satisfactory in all regards except that Kicad doesn't
build on it.  You then put the blame on Fedora?  Kicad is the dog and
Fedora is the tail?  So I should dump my entire computing environment to
buy a Mac or whatever platform seems to be stable at the moment just so
I can build Kicad every week to shuffle the bugs around? 

> kicad-install.sh has worked for some on Fedora, and as of 10 minutes
> ago, it builds the pre-kiway version which you might best think of as
> being stable.  Unless you edit the script. 

Thanks for this suggestion.  I downloaded the most recent
kicad-install.sh script and it ran to completion without any errors.  A
big improvement over my previous attempts a year ago! 

> I've given all I can afford.  You may ask more of me, but if its not
> in my common interest, it ain't coming from me. 

I'm not asking any more from you.  You do what you can, and you
have done a lot.  

I just notice that you tend to discourage other people from doing what
they want to do and telling them they are wrong-headed for trying to
solve problems that you don't happen to have from your personal view of
the world (eg: just debug and rewrite the database and recompile).  

Other people might view replacing their whole distribution (the tail) to
fix Kicad (the dog) as being slightly unrealistic.  You are very
competent, opinionated and are basically driving the development around
here.  If you decide to squash something that you don't like for bizarre
reasons, it will likely die on the vine.  Not much can stand up under
your withering attacks.  Go ahead an be an alpha dog, just try not to
bite the heads off of everyone else on the team. 

My thanks and appreciation for all that you and the rest of the team
ha've done.  I still use kicad personally and find it one of the
brightest points in the linux world. 

kind regards,
--
Rick Walker