X-Authentication-Warning:	delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient:	geda-user AT delorie DOT com
X-Original-DKIM-Signature:	v=1; a=rsa-sha256; c=relaxed/relaxed;
	d=gmail.com; s=20120113;
	h=message-id:date:from:user-agent:mime-version:to:subject
	:content-transfer-encoding;
	bh=d3+pqXgt8h5vX//mE9VPzsohMbzyja0xotUgNvh2HZ0=;
	b=E0AOeTbzKs7oYYGYaHfcqvNrSqfur3fcDRg2NXSkn3aThx99/vrZO858OxB5nYoeYN
	1GS6qdr6jBv45B/3NMQJ1OThj1/HPMeQp7VR+Lxi7zLXh8FYUm0uWaPal3JtaUjaaRN8
	1Yvcrz4/GWJnWylWLZ3dX4yReonrX6Nf+7gs1ZijsuadyQJFe6QyB6iTfQrZkmPSkO79
	L4xGUujSHRddwATfB5XEyrGj0W7iuUiUstaRO+wicGNPjF3F98vQSPdC1YuDDJ1sSrXa
	KoH1sPBdZ63QQLS09S5kWA7hig0pi9vw6MbLe5aCyU6E3whF5mdaUugTj8EK3WHm0YNE
	FkmQ==
X-Google-DKIM-Signature:	v=1; a=rsa-sha256; c=relaxed/relaxed;
	d=1e100.net; s=20130820;
	h=x-gm-message-state:message-id:date:from:user-agent:mime-version:to
	:subject:content-transfer-encoding;
	bh=d3+pqXgt8h5vX//mE9VPzsohMbzyja0xotUgNvh2HZ0=;
	b=DmxYwyH9A14qkKNCQm3oguJ2h6YVPFqEzM4PLX66l/2NBgYnoJbF3A9ZGD4pZY1TT9
	ji5bvUkekD9mIzwroZJDh4Se03SZ9wvxwvigDWpJgvoMG68rxJUL4Xlr0AB9rYiTh9s+
	Yu7ePiUyfhnTmdl6LG7sL8pH1LcHV+lLFpROrKNYyQZ3yTgjE/FI0Eyj7356/J1nGkgQ
	4VVtBjZix38CTEb0rFBoMpTJlxfYukb9+ZB47Pq1YVkodCBO/4qT21NZBOs6hPADP/Gu
	CMUv/tW8Thyihuw732s0kY9VkPQHvqQ/i2BrHAwFT3RctE8T+SYPinVkQL2fMCMnkE6n
	cn4g==
X-Gm-Message-State:	AOPr4FUMZkzKMkqNa4B+l+t2WQVUCB9dvPxx0811rz+3b8uLtQRftRgJadiCqD7QmErZcA==
X-Received:	by 10.28.175.143 with SMTP id y137mr558797wme.21.1462117144980;
	Sun, 01 May 2016 08:39:04 -0700 (PDT)
Message-ID:	<572622FE.9050705@gmail.com>
Date:	Sun, 01 May 2016 16:38:38 +0100
From:	"Iain Paton (selsinork AT gmail DOT com) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
User-Agent:	Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.6.0
MIME-Version:	1.0
To:	geda-user AT delorie DOT com
Subject:	[geda-user] gschem annoyances
Reply-To:	geda-user AT delorie DOT com

After a recent HD failure, I've just built the latest gschem git master. 
Here are some some of my thoughts from the process:

Guile 2.0.x is a real pain in the *** to build and even after getting it 
to build it interacts badly with whatever gschem is doing with libtool. 
For example you end up trying to link to a shared version of 
libatomic_ops that doesn't exist and is anyway two steps removed by 
being statically linked into the garbage collector that guile uses.
(libatomic_ops by default builds a static lib, not shared, gschem should 
neither know or care about it)

After finally solving all the problems and getting a working build I 
discover the next set of problems: Sidebars. No choice, forced on every 
time gschem is opened.

So let me describe my usage. I have a 30" 2560x1600 screen and a second 
smaller 20" monitor. In previous gschem versions I've kept the main 
schematic page on the large screen, all the menus and dialog windows on 
the second screen. This gives me the maximum useable space for the 
actual schematic I'm working on.
I'm now faced with a large, mostly empty sidebar and whatever you call 
the thing along the bottom that doesn't exactly fit the sidebar 
description. Collectively these reduce space for the schematic by just 
short of a million pixels!

Ok, so these can be hidden from the view menu, but this takes either 
four mouse clicks or four keypresses *every* time which is a serious 
blow to productivity.
Can someone with commit access please commit the patch from
https://bugs.launchpad.net/geda/+bug/1492834 so that users have at least 
some choice?

The patch isn't enough however. There's still a couple of serious 
useability problems.

Click on the 'ABC' add-text button on the toolbar, or use the AT 
keyboard shortcut to add some text and I get the pop-up dialog as before 
and this can be put on the second screen.
But if I want to edit some already existing text, rather than this 
happening in the dialog, it's duplicated in the Text tab in the sidebar.
Two problems here, if the sidebar is hidden it's automatically opened. 
This might be ok, but it has to be manually closed which quickly becomes 
annoying.
It's also inconsistent from a UI perspective to have two different 
places with essentially identical layout for no reason other than Add vs 
Edit.

Interaction between the sidebar opening and the display of the page I'm 
editing means that if I edit some text at either edge of the window then 
there's a good chance I end up with the thing I'm editing disappearing 
off the screen which isn't at all helpful.

I can also size the sidebar down to zero with the gripper. But this has 
undesirable problems as trying to edit some text then does nothing at 
all, it doesn't work out that the sidebar isn't visible and restore it, 
doesn't present the dialog, instead it just switches to an invisible tab.

The sidebar might just about be useful if it could be detached and put 
on my second screen - I can do this with both the menubar and toolbar if 
I want..  To make it worthwhile would require being able to combine it 
with other things like the Add Component window and Attribute editor 
though and that doesn't seem to be possible?

Sidebars are also completely unconfigurable - can't hide by default, 
can't move to the left, can't detach.

Bottom 'sidebar' doesn't respect (log-window "later") which I've had in 
$HOME/.gEDA/gschemrc forever to tidy up the annoyance in previous versions.

Most of what is in the sidebar appears to be seldom used, Object tab 
mostly only useful when creating symbols and Options tab for me only 
contains things that are set once and then never touched again.

On the Options menu, the Show Log Window... entry now does nothing it 
should either be updated to work with the bar, or removed.

I've added some of this stuff to 
https://bugs.launchpad.net/geda/+bug/1496899 which seems to be very much 
along the same lines as where I'd like to see this go.


Finally a request for some help. While the sidebars evolve I'd like to 
temporarily go back to the old Edit Text dialog, I've found where it's 
connected in system-gschemrc with
(,(N_ "Edit Text...")       &edit-text             "gtk-edit")
but I'm having problems tracing where that &edit-text scheme function 
ends up. Would someone more familiar with how the scheme code interacts 
with the underlying C code be able to point me in the right direction?

Iain

- Raw text -