Hi Mike/Pete et al,
Thank you very much for the notes, Mike.
On "[IPP Scan LastCall comments] TC 4: Defer until Pete and Soma can provide background; possible Note: in text": If we have a USB only IPP Scan device (for that matter, a printer device too) that does not have an RTC chip/battery that maintains clock out-of-factory (and certainly does not have network support to use NTP), it does not have a well-defined way to get/set current time. All relevant IPP standards make some form of date-time mandatory. However, a device, described above, is typically not capable of providing accurate date-time. They may provide their notion of time (which may be based on heuristics incl born-on date or simply UNIX epoch); but the accuracy of the reported date-time is unspecified. A device that supports WiFi P2P only & no RTC is in a similar position.
I believe that some guidelines to reconcile the requirement of date-time and how these devices may handle the requirement will clarify implementation options for client/server developers. This could take any of the following (or anything else suitable):
1. date-time-xxx and related attributes are conditionally required for devices with network interfaces incl devices with only P2P network connections (USB devices do not have to support these attributes)
2. date-time-xxx and related attributes is required for all IPP devices. Devices that do not have a way to maintain accurate date-time in the device across power-cycles may make their notion of wall clock time available obtained through any network services like NTP/SNTP (best effort), user set value or a value derived (i..e, counted up suitably on a timer) from a value that is arbitrarily assigned at every power-up.
3. The accuracy of the date-time-xxx and related attributes that are mandatory should conform to the specifications mentioned in section xxxxx of PWG xxxx.yy/RFC xxxx.
The notes describing PWG's position on these attributes will determine if RTC (and a way for the user to change the date-time) is required for all devices. Of course, this may mean additional cost to OEMs.
Thanks and Regards,
From: ipp-bounces at pwg.org [mailto:ipp-bounces at pwg.org] On Behalf Of Michael Sweet
Sent: Tuesday, July 29, 2014 4:53 AM
To: <ipp at pwg.org>
Subject: [IPP] Minutes posted from today's concall
I have posted the minutes from today's IPP WG conference call to:
Our next conference call is on August 4, 2014 at 3pm with the Cloud Imaging Model WG to review slides for the F2F.
Michael Sweet, Senior Printing System Engineer, PWG Chair