PMP Mail Archive: PMP> Re: Is comprise at all possible for localization? [yes!]

PMP Mail Archive: PMP> Re: Is comprise at all possible for localization? [yes!]

PMP> Re: Is comprise at all possible for localization? [yes!]

Tom Hastings (
Mon, 28 Jul 1997 14:38:40 PDT

At 12:53 07/28/97 PDT, JK Martin wrote:
>Ira and Tom,
>Trying to resolve the localization issue has not been easy.
>Despite the many proposals on the table, it appears that you
>are only interested in the proposal having the greatest impact
>on the Printer MIB.

I just talked with Ira and we like David's Friday proposal (the 6th
alternative) which uses the existing localization that is in the printer
MIB, instead of adding new char set control objects. I sent mail on Friday
to that effect as well, saing that I "thought David was on to something".

We have two suggested improvements/simplifications of David's proposal
which brings together most of the alternatives:

1. We belive that requiring an implementation to be able to change
the charset, language, and country for all of the objects that are under
control of prtGeneralCurrentLocalization is too hard. That we should
make statements that the establishment of the current locale is expected
to happen at install time (or re-install time), but not dynamically as
the printer is being used. (This seems to reflect the recent e-mail
on actual implementation as well). In other words, make the
current localization that is in RFC 1759 be more like the static
localization concept.

2. We have products that are displaying the other objects in the group
to the user, besides the prtXxxDescription, such as xxxVendorName, xxxModel,
xxxSerialNumber. Therefore, we suggest that most of the read-only objects
should be subject to the current localization: prtGeneralCurrentLocalization.

The ones that SHALL not be localized by the agent are:
R prtLocalizationLanguage PrtEnglishASCIIStringTC
R prtLocalizationCountry PrtEnglishASCIIStringTC

The ones that the management app should localize are the three objects that
do have standard text values specified:
> R/W prtInputMediaType "stationery", "transparency", ...
> R/W prtInputMediaColor "other", "unknown", "white", "pink"
> R prtMarkerColorantValue "other", "unknown", "white", "red", ...

So Ira and I think that these three should NOT be subject to localization
by the agent either (since these are values that the printer "understands").

>Are you two willing to compromise *anything* in this issue?
>In other words, if you HAD to chose one of the other proposals,
>which one would it be?

We think that David's Friday proposal with the above modifications would
be a compromise that we could all buy.

>Please state your preference as quickly as possible, before this
>issue is shutdown altogether due to lack of cooperation.
> ...jay