X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient: geda-user AT delorie DOT com
Date: Fri, 21 Jul 2017 05:30:27 +0200 (CEST)
X-X-Sender: igor2 AT igor2priv
To: "Nicklas Karlsson (nicklas DOT karlsson17 AT gmail DOT com) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
X-Debug: to=geda-user AT delorie DOT com from="gedau AT igor2 DOT repo DOT hu"
From: gedau AT igor2 DOT repo DOT hu
Subject: Re: [geda-user] [chscem] slow start
In-Reply-To: <1465D832-2CF3-44FB-961D-EFFDCFEA81D5@noqsi.com>
Message-ID: <alpine.DEB.2.00.1707210431590.27212@igor2priv>
References: <alpine DOT DEB DOT 2 DOT 00 DOT 1706300916510 DOT 27212 AT igor2priv> <20170719141700 DOT a9a156f68d8968c53ce1e46a AT gmail DOT com> <alpine DOT DEB DOT 2 DOT 00 DOT 1707191956040 DOT 27212 AT igor2priv> <CAOP4iL3v-kJ-V+n99y_9OzL5480O3KcJeEqC2tY29e6hnKOXFw AT mail DOT gmail DOT com> <f1cb7c32-99b0-03b5-f77b-5b8a24303231 AT ecosensory DOT com>
 <31B5BD5F-73B5-44F7-B1B2-19C01D7C9661 AT noqsi DOT com> <5970FE10 DOT 9080903 AT xs4all DOT nl> <1465D832-2CF3-44FB-961D-EFFDCFEA81D5 AT noqsi DOT com>
User-Agent: Alpine 2.00 (DEB 1167 2008-08-23)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
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
Precedence: bulk


(answering only because I've been addressed)

On Thu, 20 Jul 2017, John Doty wrote:

>
>@Igor2: a PCB program with a real physical model of the board interests me
>more. Go for that, as I think you?re doing.

pcb-rnd already has a rather good physical model of the board after the 
layer rewrite. At least good enough for me and most of our actual users. A 
few smaller parts are still missing, but it seems the user demand behind 
those are very low. Nevertheless, they will happen sooner or later, in 
parallel to cschem development.

I know you are happy with lepton. I am not. Some other users are not. We 
have started cschem. As Bert said: it's been decided, and it is going to 
happen. You can not stop it by pushing your "lepton is the only real 
choice and we don't need anything else" agenda. There is only one way to 
stop it: if you quickly change lepton to do _all_ that's listed in 
cschem's manifesto.

Anyway that's probably not going to happen, so you'll need to live with 
the idea that cschem is going to exist. However...

You are not affected. Cschem will not shut down lepton or github. Cschem 
will not be a virus that infects your machine and convert or delete your 
.sch files. Cschem is not a ransomware that encrypts all your precious 
scheme scripts by converting them to awk. You don't need to fear it.

Since you are making suggestions as of what I should do, I will make one 
about what you should do, even if I know you will ignore it. It goes 
like this:

Either try to be constructive about it or just stay out of it.

Repeating over and over that you don't need it will not make any change 
about the course of events about lepton or cschem. Not even if you extend 
the wording and trying to talk in the name of other users.


> Lepton will be able to feed it.

Does it already have support for the back annotation, back from 2015? The 
tEDAx netlist format? Have _you_ ever made a board using pcb-rnd and sent 
it to a fab?

I suspect the answer is no to all 3. Which means in your "I do not touch 
pcb layout ever" world lepton is the ideal tool to feed pcb-rnd. In 
reality, geda/gaf/gschem is ahead on supporting pcb-rnd - that's why 
genxproj can build on gschem but can't rely on lepton. If anyone asks, I 
will recommend them to prefer geda/gaf over lepton mainly for these 
practical reasons.

If you want lepton to be able to compete with geda/gaf/gschem on this, 
you'll need to write some code. I believe the back annotation part can not 
be solved from a backend, mostly due to a design choice I am usually 
referring to as "flaws" and going to do differently in cschem.

This does not mean I think lepton is dead. But I want to be clear about 
something: from pcb-rnd's point of view, from gexproj's point of view, and 
generally from the point of view of whole family of these new tools: 
lepton is already much less capable than geda/gaf. And unlike geda/gaf 
lepton doesn't show any sign of trying to cooperate. Just saying things 
like "we don't need anything else" or that "lepton will be able to feed 
it" while it's a few features behind geda/gaf won't help this.

I do understand that one way getting this "fixed" is to kill off the 
cooperation and progress on geda/gaf and/or cschem by trying to demotivate 
anyone who is not rolling lepton. But at least in case of cschem it just 
won't work. In pcb-rnd we managed to build a new, active and 
_constructive_ community It mostly lives on IRC with more daily traffic 
than geda-user@'s weekly traffic these days. I plan to attempt to do the 
same with cschem. That's why I took it off this list as soon as I could - 
to leave behind the demotivators.

Now you need to understand that the other way of getting this fixed is to 
follow geda/gaf and implement the missing feature (maybe even consider 
changing a thing or two in the basic design, things that I believe are 
just flaws). tEDAx should really be a 20 minutes backend task, so I am 
talking about the back annotation here.

I remember you last time said that this problem doesn't need to be solved 
- and I agree with a small addition: it doesn't need to be solved for 
_you_, in lepton. Fortunately we have geda/gaf and we will have cschem, so 
at least the rest of the world, users who do want this problem to be 
solved, already have it via geda/gaf and will have it via cschem. Whether 
you like that or not.

Regards,

Igor2