Re: gEDA-user: PCB footprint mapping

[bifferos <bifferos@xxxxxxxxxxx>]

--- Dan McMahill <dan@xxxxxxxxxxxx> wrote:
> So one of the real reasons I've never gotten around to implementing any
> sort of upload mechanism for footprints is that I've also not written
> down and published an exhaustive list of exactly how footprints must be
> done to be accepted into the main pcb distribution.  I really feel

Auditing the footprints and coming up with a perfect distribution is a 
noble enough aim, but there probably never will be enough people to do 
the auditing.  I think you need to be practical about it and include all
available footprints, categorising them into approved/unapproved sections
(or however many categories you want to have).  Suppose I'm a noob
and I choose the wrong footprint - chances are I'm going to print the 
board, and place the components on the paper before sending it off for 
manufacture.   If I get the wrong footprint I'll try again with another
one.  This is (to me) a lot more efficient than emails to the mailing
list or having to delve into footprint generation as a first hurdle 
to using PCB.  I think I would be more than happy so long as there was
some kind of indication I'm working with an unapproved footprint.

Of course the footprint has to work with the software (to load), or 
it won't be much use, but that's presumably possible to 
programmatically verify.

Don't know if you've considered systems based purely on peer review.
Users login, upload data, people write reviews, giving marks 
out of 10.  PCB maintainers syphon off symbols with high ratings
and there *are* no audits.  Democracy :-).   You see the reviews
when you browse the library.

> What else... there is the issue of "heavy" vs "light" footprints.
> You'll see some footprints in the PCB libraries where the pins have
> names like "INP" "INM" "OUT" instead of just being generic packages.
> Personally I'm opposed to including that level of information in the
> footprint.  I'd rather make the footprints generic and propagate that
> information forward from the schematic capture tool.  But, I think there
> may be some users who don't use gschem as the front end or may not even
> (gasp) use schematics.

I'm with you 100% on that.  It seems (no disrespect intended) that the
gEDA suite as it stands has enough on it's hands to meet the common 
use cases, so I'd stear clear of the less common ones.

-biff.

PS: At the risk of being shot (because I'm sure this must have been
suggested before), if the footprint was XML, it could perhaps conform 
to 'heavy' and 'light' schemas?  Also review status tags would then be 
an extension of the structure and require no software changes to support.



		
___________________________________________________________ 
To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com