Gary Scott <garylscott@[EMAIL PROTECTED]
> wrote:
> You can overwrite what has previously been sent. There is no
> requirement to unsend, only to control where the subsequent write
> begins. Terminals can do that, but more im****tantly terminal emulations
> in the form of console windows can do that. No need to write a device
> driver for every possible device. The CVF/IVF original implementation
> behaved as I would have preferred.
All of which sounds like you are looking for something that has
essentially nothing to do with any of the things that advance='no',and
for that matter, tabbing, was designed for or speced for. Tabbing around
in a record to "overwrite" previous stuff in the record essentially
*DOES* unsend, at least if you do it within a single format (i.e. before
it is physically sent). It simply is not the same thing as writing a
record and then overwriting part of it. The tabbing changes what the
record is in the first place before it is ever sent. That's what the
model of tabbing is like. Looks to me like you are looking for a
terminal control feature. You can do that (with terminal control
sequences, preferably suitably abstracted), but that's not what tabbing
was designed for. No great surprise that it doesn't do it. As in my
comment
>> I'll also
>> claim it is pretty typical of a class of debate where someone tries
>> to take a feature designed for one purpose and modify it for a
>> completely different purpose, even if that breaks its use for the
>> original purpose;
Advance='no' was for prompting and for sup****ting records larger than
the system buffer sizes. Making it act like you appear to want would
break both of those things that it was for in order to turn it into
something that it wasn't for.
I frankly don't figure it worth debating much because I'd say that the
odds of such a change getting enough sup****t to go anywhere are
negligable. I doubt the user community would accept it, and I'd be at
least slightly surprised if you managed to get a second on the committee
for such a proposal, particularly as it would be explicitly incompatible
with what the standard currently says. It tends to require an extra high
degree of sup****t to make such incompatible changes.
In general, if you want a new feature, it is a lot easier to propose a
new feature than to grab onto an existing one and propose that it be
changed to do something completely different and incompatible with what
it already does. I do see a lot of proposals like that; they tend to go
nowehere.
--
Richard Maine | Good judgement comes from experience;
email: last name at domain . net | experience comes from bad judgement.
domain: summertriangle | -- Mark Twain
|
