X-Authentication-Warning:	delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient:	geda-user AT delorie DOT com
From:	geda AT psjt DOT org (Stephan =?utf-8?Q?B=C3=B6ttcher?=)
To:	"Britton Kerin \(britton.kerin\@gmail.com\) \[via geda-user\@delorie.com\]" <geda-user AT delorie DOT com>
Subject:	Re: [geda-user] The nature of gEDA users
References:	<alpine DOT DEB DOT 2 DOT 00 DOT 1601180756390 DOT 9035 AT igor2priv>
	<alpine DOT DEB DOT 2 DOT 00 DOT 1601260416150 DOT 9035 AT igor2priv>
	<56A751EC DOT 8030402 AT iae DOT nl>
	<20160126124701 DOT 0d061912c7e078ced9d4e6cb AT gmail DOT com>
	<CANEvwqgs3YFnt7m8mA1DN6X2KdWbyr4zpXCVH321vDo1f7CyxA AT mail DOT gmail DOT com>
	<201601261804 DOT u0QI4KEQ009550 AT envy DOT delorie DOT com>
	<E7D351BF-5BBB-41AC-B996-D5E27079A82C AT noqsi DOT com>
	<CAC4O8c-ZyNnCzCDHXkYYabSD4fG8vf+CKmhMycNJujGMPKzQDQ AT mail DOT gmail DOT com>
Date:	Tue, 26 Jan 2016 22:02:01 +0100
In-Reply-To:	<CAC4O8c-ZyNnCzCDHXkYYabSD4fG8vf+CKmhMycNJujGMPKzQDQ@mail.gmail.com>
	(Britton Kerin's message of "Tue, 26 Jan 2016 11:29:07 -0900")
Message-ID:	<s6nr3h49hrq.fsf@blaulicht.dmz.brux>
User-Agent:	Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux)
MIME-Version:	1.0
X-MIME-Autoconverted:	from quoted-printable to 8bit by delorie.com id u0QL2GMl019631
Reply-To:	geda-user AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	geda-user AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

"Britton Kerin (britton DOT kerin AT gmail DOT com) [via geda-user AT delorie DOT com]"
<geda-user AT delorie DOT com> writes:

>> Pcb says to the user “my way or the highway”. Gschem says “live with
>> my quirks, and I’ll help you do anything”.
>
> Not true, I've just gone through the entire format carefully and pcb
> actually has very little per-part or per-feature overhead.  It's about
> as close to a paint program as it can be.  It can do lots of stuff,
> though admittedly often in quirky ways, just like gschem.  Your major
> gripe seems to be that you can't "draw" 3D or inter-layer features,
> because their implementation is bolted onto the fundamentally layer
> oriented design in a hard-wired way.  It's a true complaint but not
> very useful, because adding 3D CAD would require a total rewrite with
> buckets of additional complexity.
>
> It's completely possible to go on adding features like blind vias
> without a total rewrite.

I am sure, if I'd need blind vias right now, I'd do them with PCB as it
is, somehow.  It would not be beautiful.  Just did a rigid-flex where
the checkout produces three copies of the PCB file via AWK to export
various layers from each.  One copy edited the layer grouping to move
some pads from layers 1-6 to layers 3-4.  And I needed to remove the
microscopic vias that connect those connectors pads to layers 3-4 whle
they where still on layers 1-6, for connectivity checks.

I'd love to do these things natively in the file format.  It does not
even need to have much support in the GUI HIDs initially.

I believe PCB can be developed to support all this.

But the first steps should not be "implement b/b vias" or "allow more
than one pair of outer layers".  It should be something like "expess
Elements and Vias as subclasses of generic containers that can include
anything".  It should be steps that allow to do anything that may come
up in the future that can be expressed in (Gerber) layers.

I believe that 2D layers are and should remain the core concept of
PCB. For b/b vias, holes should get their own layer(s).  There can be
more than one of those, and each has attributes that say which copper
layers they connect to.  Via stacks would be composed of circles on a
hole layer and circles on copper layers, as required.

That is all that is needed at the core.  Adding GUI support is the next
step, and an entirely separate problem.

-- 
Stephan

- Raw text -