> > While most Unicoders like to think that terminal emulation is is a relic
> > of the past, there are numerous terminal emulators on the market and in
> > active development, and millions of people who use them for applications
> > ranging from software development, to email, to transaction processing, to
> > system administration.
> Of course this is true. However... How many of those emulators and people
> really require these particular few DEC VT100 graphic codes? Will their
> usage increase over time?
I don't think that's a criterion for including a character in Unicode.
> I think that if Digital originally didn't bother with them, they're better
> left alone.
Digital is famous for killing off products against the will of its customers,
and ultimately for killing itself off.
As a maker of terminal emulation software, I can tell you that the issue of
horizontal scan lines comes up from time to time, as do requests for other
characters from other terminals that are not presently included in Unicode,
numbering in the hundreds. The people who raise these issues don't always
need to be told what is best for them.
> > Each maker of terminal emulation products does the same thing, but does it
> > differently, and this is a rather shameful and unnecessary waste of time
> > and labor, and one that inhibits interoperability of applications.
> Then they should band together and get themselves organized so they can
> request the addition of their required graphical segments and line-drawing
I hope that will be the outcome of this discussion.
> I don't believe there's that much interest.
Perhaps not compared to today's hot mass-market items, but "niche" markets
are often important areas where real work is done. If all we cared about
was how much interest there was in a particular product, we'd only care
about one product.
> > Terminal emulation will be with us for years -- probably decades -- to
> > come.
> Of course that's true. It'll just get less like "dumb terminal"
> emulation. I don't see a big, continuing market in the 24x80
> one-column-one-char terminal emulator. However, I do see a great
> long-term need for the command-line interface and interpretive languages,
> such as shells, LISP, etc. Those just don't require the line-drawing
> legacy stuff.
The degree to which terminal-host communications are used for data entry is
easily forgotten by those who read the popular computer press. However,
countless people do indeed require full and accurate emulation of specific
terminals, including their forms-filling and math / technical features.
Forms filling is especially important in mission-critical
transaction-processing applications, and when this requirement is combined
with the requirement to enter accented and/or non-Roman letters, let alone
Kanji, etc, the need for Unicode support for this environment is evident.
Meanwhile, perhaps contrary to expectations, the need for accurate terminal
emulation grows in this market segment, as actual terminals -- which are
indeed rather hard to come by these days -- are replaced by PCs.
> I use a terminal emulator all the time, daily -- I almost
> never need even a rudimentary cursor positioning interface, and *NEVER*
> need a line drawing interface to it.
This tends to be true of people who are involved in development rather than
production. Makers of terminal emulation software, however, know the
demands of their users, particularly users outside the English-speaking
world, and particulary those whose applications need more than just cursor
positioning. A real VT220 (for example) allows you to type Portuguese or
Norwegian names and addresses into a complex form; when the terminal dies and
is replaced by a PC, the customer expects no less.
Anyway, I hope nobody will mind too much if I post a longish draft proposal
on this topic shortly, in hopes of getting some constructive commentary and
additional material, before submitting a formal proposal.
Meanwhile, I would like to begin with a more modest proposal: that the word
"legacy" -- an emotionally toned would-be self-fulfilling prophecy coined by
mass marketeers to get your money -- be left out of this discussion. What
is "legacy" to one person is stable, efficient, dependable, simple, and
usable to another.
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:42 EDT