In message <[log in to unmask]>, "Senn, Bruce"
<[log in to unmask]> writes
>FWIW,  I've often thought of the difference between RETURN/ENTER as a kind
>of basic test for computer users.  If you can't recognize the difference in
>what the keys do, then, well, maybe, computers aren't for you.  ;-)
>
>I think I'd rather have them separate.
>
>Bruce.

It is quite a test for the *programmer* to figure out what the
difference is...

As for the user, Return and Enter are the last vestiges of a
hardware-centric approach to the HP3000 that began with the HP264x
terminals, where you actually had to push a button to put them into
Block Mode; it couldn't be done programmatically.

And Return and Enter reflected two very different ways of managing the
buffer. Return sent a character that said to the HP 'Hey, remember that
data I sent you?' Well now you can act on it'. (Because in character
mode, the data string to be acted on had already been sent, a character
at a time, and Return said 'Go'.)

You didn't actually need the Return if you had a fixed length of input;
once the HP app had that many characters, if could start on its own.
That had its uses, but wasn't universally popular...

While Enter doesn't send a character at all, not per se; it's a trigger
for the terminal to send its buffer. Fine in block mode, but it causes
the strange 'double input' feature in character mode, as the buffer it
is sending is already sent.

But what has *any* of this to do with the user? They press a button, the
stuff goes, they don't care about the internal workings. Why should
they?

'Oh, you're in block mode, you have to press this *other* Enter key'. A
big barrier to the transparent use of the HP3000, and a throwback to the
days when users had to adjust to the computer and not the other way
round. Says old, says klunky.

Even experienced HP3000 users get confused by apps that switch. Mix a
nice block mode manufacturing system with local Powerhouse add-ons -
'uh-oh, gotta remember which is which'. Interoperability issues *on* the
platform, let alone with other platforms.

It's been a bee in my bonnet since the early eighties. I tried to get HP
interested in 'Return = Enter in Block Mode" as a terminal setting, when
they started doing terminals where you could set 'Return=Enter', albeit
not flexibly enough. No go.

I tried to get WRQ interested in it, as a USP for Reflection. No go
there either.

Finally, I got the chance to work on a new terminal emulator, ScreenJet.
The bit where Return could be reconfigured wasn't my area, but I talked
to the guy whose area it was, and we got proof of concept in an
afternoon. A few refinements, and we were away.

To me it was critical to the future of the HP3000: if using this
platform is to be a no-issue no-brainer (and maybe users aren't even
supposed to know or care they are on an HP3000, except that it doesn't
keep going down, like their competitors' systems), you can't expect
users to learn 'Return here, but Enter there'. Period.

So I loved the way it kept creeping higher and higher up the feature
list in ScreenJet literature. And how ScreenJet's competitors started
practising that sincerest form of flattery (because, hell, we *all* want
the HP3000 to have the best possible chance of survival, so we all need
this advantage).

And now to be the default in QCTerm, even though it's more like what
it's emulating if it isn't the default. Because this makes it *better*
than what it's emulating..... gives me a warm glow....

(Oh, and it makes laptops *infinitely* more usable as terminals - even
if you know the F12 kludge)...

Go, Wirt!
--
Roy Brown        'Have nothing in your houses that you do not know to be
Kelmscott Ltd     useful, or believe to be beautiful'  William Morris

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *