Mail Archives: geda-user/2017/02/16/22:07:28
On Thu, 16 Feb 2017, Dave McGuire (mcguire AT neurotica DOT com) [via geda-user AT delorie DOT com] wrote:
> On 02/16/2017 09:03 AM, John Doty wrote:
>>> We currently have 4 gEDA projects that will need to interact in random
>>> ways:
>>>
>>> - two variants of gschem/gnetlist/geda/gaf
>>>
>>> - two variants of pcb
>>
>> You left out Icarus: it?s also a gEDA project. And then, gschem is the
>> schematic capture of choice for many ngspice users. There are many less
>> trodden paths: ?gnetlist ?list-backends? will show you some. gEDA is much
>> broader than your vision.
>>
>> You complain of politics, but when you leave users out of your vision, you
>> are engaging in politics.
>
> Much like you leave entire (large!) classes of users out of your vision.
>
> -Dave
>
Just to clarify, if we do not take my original sentence out of context: I
was describing a single problem affecting exclusively the sch -> PCB flow.
From the sch -> PCB flow I DID exclude icarus, spice, fortran compilers,
the travelling salesman problem, purple elephants, John's famous
satellites and his various road vehicles. (I DID include John's famous
toolkit agenda, tho.)
Facts: from the tEDAx idea I DID NOT exclude any non-pcb flow:
- while I first wanted to provide a solution for this specific problem we
already have, to kick off the project, and do it fast, I DID include
provisisions for other flows IN ADVANCE
- John either did not read any of the spec or did not find that part in
the incredible sea of text (3.7 kilobytes). Or he didn't understand that
"different backend software" can mean icarus too. Or he didn't understand
that "e.g. PCB editors, simulators" may include spice. (Quotes from the
first 500 bytes of the netlist spec of tEDAx).
- I then explicitly and publicly asked people (NOT excluding John) to
join and help me improving especially that part, the non-sch-pcb-flow.
- John did NOT sign up to help (surprise, surprise; btw, another person,
who made big contribution to pcb-rnd 1.2.1 too, did contact me with his
tEDAx+spice ideas.)
- instead John randomly took out small parts, even single sentences from
a system of domain specific reasoning and tried to apply them on fields
they never meant for. Then tried to explain how I should intrepret my own
thoughts, because he knows better what I really meant when I used a word.
So business as usual. There's no solution for this here. There might be
workarounds, tho:
Don't talk, code.
Don't complain, help them making it better.
If they don't want that, make your own, and make that perfect.
Ignore those who are clearly not going to do anything useful but are
actively trying to block everyone and everything. Ignore them even if they
build satellites or did contribute a decade ago.
Anyway, to get back to the real topic...
I hope the parties involved in the sch -> PCB flow will sit down and read
the proposal in details and evaluate the costs & benefits. I hope they
will decide to go for it. More specifically, I hope they will decide to go
for IMPLEMENTING it, and not for fighting over it.
I don't think we should be mailing 8..10 hours about this (and John's
alternative single-person universe) when the whole initial design +
specification + implementation took less then 4 hours to make.
I've already done my part. Now I selfishly lean back and wait to see if
other parties are willing to join and do theirs. Not on the mailing list
as talk, but in their software as code. I also hope this will happen
outside of gEDA. Who knows, maybe even sooner than within gEDA - it's up
to us.
(I think I will generally be more selfish regarding these topics: I will
probably ignore or shortly dismiss ideas/opinions that I don't
see directly contributing to our ecosystem, to tEDAx or to pcb-rnd. Even
such cheap/simple proposals end up being a waste of my time spending 2x
more time on mailing on geda-user than doing the actual code.)
Regards,
Igor2
- Raw text -