On Oct 19, 2015, at 11:30 PM, Britton = Kerin (britton DOT kerin AT gmail DOT com) = [via geda-user AT delorie DOT com] = <geda-user AT delorie DOT com>= wrote:

On Mon, Oct 19, 2015 at 5:41 PM, John Doty <jpd AT noqsi DOT com> wrote:

On Oct 19, 2015, at 5:40 PM, DJ Delorie <dj AT delorie DOT com> wrote:

>
>>> My sample schematic would have U1-1 and U1-1.
>>
>> So you don't really have a refdes to start from: it doesn't = identify
>> the component.
>
> Wordplay. It has a refdes. It's not unique. We do = this all the time
> for slotted and multi-sym parts.
>
> And that reference was to show how easy it is for a user to
> instantiate symbols that don't have a unique identity, a problem = you
> still haven't addressed.

It=92s easy to do all kinds of dysfunctional things with a power = tool. To reliably prevent dysfunction, you have to take away the = power.

You sound like a = surgeon resisting a checklist. =

Not at all. I=92m = one of the minority who use gnet-drc2 despite its = flaws.

"Sure I sometimes take out the wrong kidney, BFD I'm = smart I don't need that". Power tools shouldn't gratuitously force = you to remember things a competent tool can easily remember for = you.

And the toolkit = should have a way to *advise* you. Not *force* you to conform, as DJ = wants.

= If refdes aren't gonna be unique then some other thing should exist = that makes obviously distinct symbols = unique.

Yes. And = since gschem doesn=92t have complete knowledge of the environment in = which the schematic page will be used, it can=92t be the = enforcer.

I don't remember what ugly thing gEDA did to me with = duplicate refdes but I remember it being bad enough I immediately added = machinery to my setup so it couldn't happen = again.

You can detect = these with gnet-drc2, although it=92ll complain about a lot of things it = shouldn=92t. That=92s a fine illustration of the problem of developers = making assumptions about how users will use the tool. Gnet-drc2 makes = many unreasonable assumptions. The good news is that gnet-drc2 is only = an advisor, so you can ignore it when it=92s wrong. It would be = catastrophic if we forced this stuff on = people.

The trouble with this kind of thing is that = it=92s often hard to completely disable. Developers make unwarranted = assumptions. It=92s better to have it as an optional script. It might be = nice to have a gnetlist -m script to work with the tamer back ends in = one pass for straightforward designs.