gEDA-dev: [RFC] Embedding/archiving compromise

[John Griessen]

John Doty wrote:
> On Oct 20, 2007, at 1:40 AM, Peter TB Brett wrote:
  I don't think anyone has disputed that this is a good idea.
> 
> Nobody disputes this because hardly anybody uses embedded symbols in  
> active designs, I think. They are broken: only good for frozen export.
> 
> If you fix *that*, your proposal still leads to confusion. Right now,  
> a symbol edit has the scope of the library that contains it, but now  
> you'll have edits whose scope is a single file. But usually, you want  
> the the scope to either be:
> 
> 1. The whole project
> or
> 2. A single symbol.

Yes, I agree.  In the future we can evolve this set  of tools to handle any scale
of design well if we model it after what is successful in chip design...
referenced objects that can be schematic or layout or model.  Gschem and PCB ultimately can merge
and use gnetman as the data engine for symbols, footprints, schematics, layouts. assembly drawings,
marketing brochures even, all searchable by
project defined
library search paths.

Embedded symbols in files is a block to this future possibility -- it bogs down as you go to
large scales of 500 different schematic chunks.


Project defined
library search paths are a good thing.

John G
-- 
Ecosensory   Austin TX