["Bruce Sherwood" <basherwo@unity.ncsu.edu>] Re: [Idle-dev] new idlefork package

[Kurt B. Kaiser]

-------------------- Start of forwarded message --------------------
Received: from (relay2.primushost.com) [207.244.125.21] 
	by nautilus.shore.net with esmtp (Exim)
	id 18UD4m-0003Te-00; Thu, 02 Jan 2003 16:40:28 -0500
Received: from ncsmtp03.ogw.rr.com [24.93.67.84] 
	by relay2.primushost.com with esmtp (Exim)
	id 18UD5j-0007MH-00; Thu, 02 Jan 2003 16:41:27 -0500
Received: from mail4.nc.rr.com (fe4 [24.93.67.51])
	by ncsmtp03.ogw.rr.com (8.12.5/8.12.2) with ESMTP id h02LdkiZ013953
	for <kbk@shore.net>; Thu, 2 Jan 2003 16:39:46 -0500 (EST)
Received: from hyperon ([24.74.143.200]) by mail4.nc.rr.com  with Microsoft SMTPSVC(5.5.1877.757.75);
	 Thu, 2 Jan 2003 16:41:15 -0500
Message-ID: <001f01c2b2a7$929f6f40$6401a8c0@hyperon>
From: "Bruce Sherwood" <basherwo@unity.ncsu.edu>
To: "Kurt B. Kaiser" <kbk@shore.net>
References: <002601c2b14e$9d48e200$6401a8c0@hyperon> <m3bs2ze03q.fsf@float.attbi.com>
Subject: Re: [Idle-dev] new idlefork package
Date: Thu, 2 Jan 2003 16:40:30 -0500
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 6.00.2720.3000
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000

[Kurt B. Kaiser]
> > I realize that many experienced programmers are uncomfortable with
> > unannounced saves, but with the infinite un-do of idlefork this
> > really isn't a problem, and you can always save a copy if you wish
> > at any stage. High interactivity is the vitally important issue
> > here. It isn't just the need to hit another key to get past the
> > dialog box; it's the added cognitive load of an interruption to the
> > work -- you lose focus needlessly.
>
> I'd pretty much agree.  The problem I see is hitting F5 when your
> window focus was in the wrong window.  Of course, you'd get an
> unexpected run of the "wrong" module, but the user might not see the
> little * disappear from the window title bar.  Should we pop up a
> "Saving foo.py" warning dialog for one second and then proceed (with
> no further acknowledgement)?

I wouldn't want any dialog box to pop up to break the train of thought. The
only warning I can imagine possibly being okay would be on the first
autosave to say something like "Saving foo.py, as will be done automatically
on each run." This wouldn't show again, but it would alert everyone that
this is how the environment works (assuming the preference has been set to
autosave = on). The possible value of this one-time warning is that many
users (including students in a class) may not be the ones who set the
preferences and are therefore unaware that autosave is in effect. In fact,
in many institutional settings the user CAN'T set the preferences, because
Python is in a read-only directory. (Hmm. This makes me realize that
eventually there may be requests for a user's own preferences to override
those in the idlefork directory.)

> Auto save should be a configuration option; I don't know if I can get
> to it before the merge because it means further development of the
> Settings dialog.  But I can hope.  Could you post this as an RFE?

I'll do that.

Bruce Sherwood



-------------------- End of forwarded message --------------------