Re: Telnet server multiple instances?

Hello Jim,

Am 12.05.2020 um 16:58 schrieb <midrangel@xxxxxxxxxxxxxxxxx> <midrangel@xxxxxxxxxxxxxxxxx>:

Ah yes, more of a hobby machine. Remember that folks like Larry and I are always thinking that systems are production, even development systems are production to us, since developers are our users as well. So our answers will almost always be pointed to that set of needs. So we don't mean to be harsh, just realistic.

Duly noted. I know about habits, so I try to also not be harsh and explain once again, that I'm one of the black sheep here in the group. ;-)

I honestly don't know of a way for the system to determine which sign on panel to present to a user, the potential exception of the inbound IP address for the exit program.

That's sufficient for me.

Between myself and Larry (along with Paul, and others) we've been fighting that specific topic off and on for over 75 years (collectively) and we not have found a suitable method of achieving the goal you have without the user providing some level of information first. I suspect that if we ever come up with a method to do it, IBM would embrace it and assist, however we have thus far failed to be clever enough to do it.

I don't know if IBM would bother to extend the 5250 signon-screen or even protocol nowadays. My proposed solution would be what you suggested earlier: The client shall supply the intended language.
Maybe it's sufficient to send the code page (273 = German) back in the data stream, so there'd be no additional configuration necessary. "Just" new clients needing to understand the extension. Or, really configure the wanted language (2929 = German, 2924 = US-English, etc.) in the emulator and send this.
If there's no language "tag" received from the client, use the system default language. If the secondary language requested by the client isn't available, use the system default language.

If there's no way for a 5250 protocol change, I'd propose to use some fancy scrollable selection section on the signon screen, providing a list of installed languages (primary and secondary) according to ISO 3166 and/or ISO 639. Maybe adherance to RFC 5646 might be a good thing. I don't know how and if this proposal could probably work with non-latin character based languages (Russia, Arabic and most of the Asian Region). I wonder if Booth could come up with a fancy design. :-)

:wq! PoC

PGP-Key: DDD3 4ABF 6413 38DE - https://www.pocnet.net/poc-key.asc