T.R | Title | User | Personal Name | Date | Lines |
---|
2026.1 | pointer.. | SUBWAY::GRAHAM | if ya want home cookin, stay home | Sat Feb 03 1990 23:47 | 8 |
|
>But exactly how do I (or more importantly a customer) go about
>building a DECwindows Server Extension for ULTRIX?
You can try gilroy::(chris)kent for some ideas - how
the Display Postscript extension was done.
Kris..
|
2026.2 | Customers *don't* write DW Extensions ;( | GOLLY::MILLER | I need 'Deeper Understanding' | Sun Feb 04 1990 06:33 | 27 |
| <<< Note 2026.1 by SUBWAY::GRAHAM "if ya want home cookin, stay home" >>>
-< pointer.. >-
Thanks Kris. I was curious about Display Postscript, but didn't know
who to contact.
I got in touch with Product Management gilroy::(john)hall and found out
that it's basically "you *can't* (very easily)"!
So...I submitted Phase 0 requirements for V4.next to support this.
Customers and layered products (like us) are just gonna *have* to have
this in the future if they're gonna have any hope in supporting
multiple DECwindows hardware/software!
P.S. Turns out it's not even terribly easy on VMS since you have to
rely on structures (client struct in particular) that are typedef'd in
header files which aren't normally provided. Thus, if these change out
from under you (like they just did between VMS V5.3 and V5.3-1), you're
suddenly incompatible!
Regards,
== ken miller ==
DEC/Test Manager Development
|
2026.3 | | DECWIN::FISHER | Burns Fisher 381-1466, ZKO3-4/W23 | Mon Feb 05 1990 13:04 | 7 |
| Just fyi, one of the reasons that customer-writable extensions are HARD
in both VMS and Ultrix is because we are between the rock and the hardplace
on the interface. On the one hand, we want our interface to be similar to MIT
to enable us to get extensions that appear there. On the other hand, MIT refuses
to define a stable extension interface, so we can't document one for ours.
Burns
|