Backspace sends Delete by default; should send 'backspace (control-h)

[GoogleCodeExporter]

What steps will reproduce the problem?
1. In new version, I go into my editor to edit, and all of a sudden it
thinks I'm on a braindead Linus-approved Console.  Backspace has never
equaled delete.  There *used* to be keyboards where a "rubout" that was
equated to 'del' was put there, but NEVER on the PC has that been the case.
 People confuse the Backspace key with Rubout.  There was a semantic break
when they changed names.  Backspace was meant to backup and delete the
character before the cursor -- not the character under the cursor as
"rubout" would indicate.

If someone has a keyboard that says rubout -- they let them configure it to
be 'Del', but all the keyboards in the US (I can't speak for the world, but
I thought this was true for all PC keyboards), have Backspace now, and no
longer have RUBOUT.  If someone wants the same functionality as RUBOUT now,
they should be using the 'DEL' key.

But please, don't follow the old-time hardliners in their mistaken
"protest" at backspace replacing rubout.  They are different keys with
different semantics.  

I would sincerely urge you to make the default 'control-h', which on the PC
(which is where this program runs -- does it run some place else?)  is the
key that people expect.

You need to look at your *target* audience -- PC-users who are used to
using the keyboard on the PC and used to the PC keyboard behaving a
specific way.  To change that default it just going to cause confusion and
is really causing people unnecessary confusion and/or pain.  (Besides my
assertion that those who want their 'rubout' key should move into the
modern age.  Linus included...but they think it is fun to enforce such
nonsense upon the masses, so they do.  But it is NOT user friendly and is
certainly yet another example of why linux hasn't grown more in the market
place and hasn't taken off as a consumer platform -- it's because the
dinosaurs among them hold to purity of purpose rather being user-friendly.
 Problem is, they are asking the world to adapt to them, rather than
providing tools that optimize the user's potential.  Instead the users must
waste their potential while they adapt to an non-intuitive and awkward
system thrust upon them by zealots.  Grr....

Please consider what I've said and go the user-centric route -- don't go
the route of forcing users to adapt to ill-thought out quirks.

Linda Walsh

Original issue reported on code.google.com by min...@tlinx.org on 6 Nov 2009 at 7:33

[GoogleCodeExporter]

I'm sick and tired of this topic. Please use the option on the Keys page to 
change it
to ^H.

Original comment by andy.koppe on 6 Nov 2009 at 7:40

• Changed state: Rejected

[GoogleCodeExporter]

Sorry, I am unfamiliar with discussion that has come prior to this.  This is my 
first
comment on the topic.

I had the earlier version installed, and installed the new version only to see 
the
default change.  

I wanted to enter my reasoning for keeping it as it was -- this is a product 
running
on Windows, not on 20 year old unix (and the key board isn't the same keyboard).

What reasoning was given for making back-space, (control-h), into a rubout?  
control-h is considered a backspace while 'DEL' is considered a rubout in the PC
world.  What logic did people use to convince you otherwise?  What was wrong 
with the
way it was?  I.e. - Default to PC-standard, and provide and explicit switch to 
set as
equiv to control-h or delete.

Sorry you are sick of this.  I didn't know.  But it also doesn't mean the 
current
default is that which is least likely to cause confusion for PC users.  Where 
is this
product "delivered" -- who are the users?  Is it available on any platform 
other than
the PC?

I don't know of any programs that expect 'Backspace' to be mapped to Del.

-linda

Original comment by min...@tlinx.org on 8 Nov 2009 at 12:46

[GoogleCodeExporter]

The reasons were:
- ^? makes Backspace distinguishable from Ctrl+H (primarily for help in emacs)
- ^? is the default on Linux and Cygwin is aiming for Linux compatibility
- most people connecting to remote systems would be connecting to Linux boxes

You're probably quite aware that there are historic arguments for ^? too, but 
I'm not
gonna get into that, as it's about as fruitful a discussion as the 
little-endian vs
big-endian thing. The vast majority of users couldn't care less what keycode
backspace sends, but whichever one I pick, a vocal minority is gonna be unhappy.

Original comment by andy.koppe on 8 Nov 2009 at 6:52