AT&T 630 terminal - software ??

[Dan Ellard]

>From article <9434 at ihlpb.ATT.COM, by gregg at ihlpb.ATT.COM:
> Besides, if you think that your AST/286 is so good, run the following
> C program, if you get the output
> 
> 1
> 70000
> 
> Your C compiler wins my award for effort otherwise you won't get any
> words of sympathy from me...
> 
> main ()
> {
>         unsigned long i;
>         char *s, *malloc();
> 
>         s = malloc ((unsigned)70000);
> 
>         for (i=0; i < ((unsigned)70000); i++)
>                 s[i] = i;
> 
>         printf ("%d\n%d\n", s[1], s[(unsigned)70000]);
> }

I have a funny feeling this program probably won't generate that
output on any machine.  The main loop stops at 70000 - 1, and
since you only malloc 70000 chars, s [70000] might not be a valid
address anyway....

But seriously, I don't think there is any question about whether
or not an increase in the number of windows or an increase in the
number of lines in each window significantly increases PROGRAMMER
productivity-- it does.  I find having a bit-mapped screen also
reduces the amount of time I have to spend doing document preparation--
I can preview troff or TeX output at my desk, instead of having to 
hike to the printer room and back, it's a big win.  (programmers
spend a lot of time writing documentation, right? :-|)

I think Stephen Friedl's estimate of $100k a year to support a
programmer may even be a bit low, which means it doesn't take
many percentage points of increased productivity to justify a
workstation on each programmer's desk.

On the other hand, it just doesn't make sense to put a workstation
(or the same kind of workstation, for that matter) on everyone's
desk.  Some tasks, such as data entry or reading email, don't currently
benefit from using anything better than a dumb terminal, and perhaps
they never will.

Dumb terminals are here to stay.  So are PC's and workstations.  Each
has its own niche, where it is the most efficient thing to use.

-Dan

Dan Ellard -- ellard at vax.bbn.com