27 Jul
2025
27 Jul
'25
8:16 p.m.
On 2025-07-27 20:10, Paul Koning wrote:
I would not do DECdns. Overcomplicated for something that is essentially
just Phase IV anyway.
Either just use NICE, or else clob something super simple together.
On Jul 27, 2025, at 2:03 PM, Johnny Billquist
<bqt(a)softjar.se>wrote;wrote:
On 2025-07-27 20:00, Robert Armstrong wrote:
...
:-)
Of course the historically most proper way to do that would be to implement DECdns.
That's a spec I have. Then again, I also have the DECdts spec and haven't done
that one yet, and that's a lot simpler. Thanks!
I'm just waiting for the day when Paul implements dynamic name lookupsin PyDECnet.
;-)
Point to some machine, and then just pull names from there when neededmaybe?
Johnny DECdts has a mode that uses DCE as a transport, and
I'm not sure what DCE looks like. Does anyone know where to find specs? I sort of
rememberit being an industry standard.
I wouldn't go there either. Probably do something very simple, maybe
with some inspiration from NTP, but primarily just for getting a roughly
correct clock without any expectation of sub-second accuracy. With
simulators this gets pretty pointless anyway.
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt(a)softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol