Re: tn3270, mset and map3270

[der Mouse <mouse%Rodents-Montreal.ORG@localhost>]

> Nothing prevents a serial-attached terminal from sending key-up and
> key-down codes of whatever kind it prefers.  Nothing except of course
> legacy Unix-derived system software that can't cope with the concept.

Not so much Unix-derived; other OSes also tend to expect serial lines
to carry characters rather than keystrokes (and, like Unices, could
probably be hacked on to take keystrokes, given reason enough).

A stronger reason, I suspect, is history, with interoperability a close
second.  But the reasons don't really matter much, since, unless you're
willing to completely abandon compatability with existing serial-line
devices, you will be getting characters rather than keystrokes.

>> [...]
> [...]

Shrug.  You asked for examples, I offered mine.  Whether you consider
it worth paying attention to is up to you.

As for your doubt "whether either of [the functionalities that would be
casualties of a switch to curses] really matters" - at least some of
the application-specific optimizations very much do matter; they take
performance from unusable to usable in some of my use cases.

>> Furthermore, it's pretty pointless.  If I as a code author were
>> subjected to such a system, honestly, I would most likely simply
>> ship a copy of libtermcap and/or libterminfo with my application.
> ...which would accomplish no more than redboxing on a tcpip network.
> :-)

> (red? I think I mean red...)

Wikipedia leads me to think redboxing is a form of coin-phone fraud
based on forging in-band signals.  If that's what you mean, I don't see
the similarity.

/~\ The ASCII                             Mouse
\ / Ribbon Campaign
 X  Against HTML                mouse%rodents-montreal.org@localhost
/ \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B