Re: DBESCWT

Peter Lancashire wrote: > June Tong wrote: > > > > Peter Lancashire wrote: > > > > > A quick look in termcap shows that Wyse 50 sends Ctrl+A followed by > > > another character for its function keys. I had terminals that did this > > > (Fortune). This will foul up your field editing as Ctrl+A toggles > > > insertion. > > > > > > As you cannot override Ctrl+A but you can override the ACCEPT key (ESC), > > > I suspect you will have fewer problems redefining ACCEPT to something > > > more than ESC, or even use a separate control key, as June suggests. > > > > Or just replace all the other Function keys with something else. The ESC as ACCEPT > > is sort of problematic, in that the delay between the ESC and other parts of the > > sequence causes 4GL to interpret the ESC by itself as an ACCEPT, and do weird stuff > > with the rest of the sequence when it arrives, but if you stopped the users from > > pressing the Function keys which are causing this problem, e.g. by replacing all > > those Function keys with Control-key equivalents, they would no longer have this > > problem. > > > > June > > -- > > june_t@hotmail.com > > Grounded in Palo Alto, living on Oreo's > > Yes, that would work from a technical standpoint. From a user interface > standpoint, I like to be able to issue my users with a little strip of > paper to go across the top of the keyboard where the function keys are. > That way they know what each (single) key press will do. Would you > prefer a single keypress to a double one? Ask some users... > > Now if only I could find a way to label the fronts of all my keys > without the little bits of paper falling into the keyboard... :-) As the one who started this all (and I thank you all for continuing this discussion) I thought I ought to chime back in. 1. As some have pointed out, Art's suggestion about switching terminal type just shifts the problem. Most terminal emulations send multibyte sequences for the function keys, and thus the problem would still exist, albeit in a different form. 2. Changing the accept key would be a real user education issue. With about 650 users in 25 countries, retraining them in a data entry intensive application NOT to use escape would be challenging to say the least. Not to mention all of the documentation changes on user prompts that would have to be changed as well. 3. As for June's last suggestion of using control keys instead; finding 10 unbound control keys that are available in every application would be equally challenging impossible. Between the ones already mapped by informix (like control-a, control-x, etc.) and the many we already use, it might not even be possible. The big question we have is -- why is this just starting now? Users have been on this X.25 network for 3-4 years. Furthermore, why has changing the DBESCWT to 2 seconds not resolved the problem? Unfortunately, these are no longer Informix questions with concise answers. Thanks again for your help. Gregory King