Michael,
> On Jun 21, 2024, at 5:06 PM, Michael Ziller via ipp <ipp at pwg.org> wrote:
>> But if the user connects using the IP, 192.168.1.37, are they really expecting the print queue name to show up as “192.168.1.37”?
> We have precedent for this for TCP ports in Windows and it queries the device for the queue name to use and displays “Manufacturer Model” just like with DNS-SD/mDNS.
> Wouldn’t it make sense to use printer-dns-sd-name here as well then?
The DNS-SD name's scope is (generally) the printer's local subnet and so might conflict if you are accessing it from a different subnet, which might be the case if you are using an IP address or hostname instead of finding it via DNS-SD lookup.
The recommendation I am thinking of putting in IPP Everywhere (subject to review of course) is that "printer-name" defaults to "MAKE MODEL (SERIAL-NUMBER-OR-MAC-ADDRESS)" and "printer-dns-sd-name" defaults to the "printer-name" value. Thus, "printer-name" retains the semantics of being the user friendly name to identify the printer with and "printer-dns-sd-name" tracks the current DNS-SD service instance name with any conflicts resolved.
We can also talk about how the other IPP attributes ("printer-geo-location", "printer-info", "printer-location", "printer-make-and-model", and "printer-serial-number") are presented/used, with examples to clarify the semantics of each.
________________________
Michael Sweet