Re: [Fwd: [evolution-patches] addressbook fixes]

[William Jon McCann <mccannwj pha jhu edu>]

Christian Rose wrote:
> tis 2004-09-28 klockan 16.22 skrev William Jon McCann:
> > Here is an updated patch.  It is similar to the one that has been 
> > applied to head.  It fixes a missing default value and removes spurious 
> > debug messages.
> >
> > It differs from the one on head in that it doesn't change existing 
> > strings and only adds one string to the schema.in.in file.  This is done 
> > by using long descriptions that duplicate the short descriptions.
>
> Regarding the reuse of strings in schemas.in: If all you do is reuse a
> string that is identical to one already marked for translation
> elsewhere, then from the translator POV it's not a new string, since
> existing translations will be reused automatically. So those need not
> string freeze approval.

Good, that was my intention. :)

> Regarding the new added string, "Show the \"Preview\" pane", I'd still
> like to know what the problem is that it fixes, and how important that
> fix is considered to be. If it's not terribly important, then it should
> probably only go into HEAD, but if it fixes some kind of problem that
> developers consider important to fix in GNOME 2.8, then it's a different
> thing. But I'd really like to know for sure (is there a bug report about
> the problem somewhere?) and some more details before deciding.

The schema for the preview pane visibility GConf setting is missing. 
This means that the default value is not set.  In this case, 
gconf_client_get_bool will return FALSE.  The "Preview" pane is really a 
misnomer, it is, in fact, the "View" pane.  So, by default, the user 
will not see Addressbook entries when they are selected.  This is very 
confusing for first-time users.  Most will not think to look in the view 
menu for an option to turn something on that they don't already know 
exists.  Also some others don't think the "Preview" pane describes what 
they need.

In fact, I only discovered this bug after doing some first-time user 
testing of Evolution-2.0 on RHEL3.  It was one of the most common 
complaints (many of the others have already been fixed).  This hits the 
user in that critical initial evaluation period.

Jon