Re: FFMII Identifier value space

[Johannes Lehtinen <johannes.lehtinen@rossum.fi>]

As a continuation to the unification of identifiers issue...

I have identified following locations in the spec where all lower case identifiers are used.

8.1.3 Standard capabilities

Capability identifiers, e.g. "wrm", "rdm.users" would become "WRM", "RDM.Users" and so on.

Property names, e.g. "write-protected", "restricted-read" would become "WriteProtected", "RestrictedRead" and so on. This is also better in line with how class properties are named.

8.2.1.3 Actions

In wrm.ActionSpecification property GenericType: "accept", "reject" would become "Accept", "Reject" and so on.

in wrm.ActionSpecification property NextStepId: The special reserved value "_pop" could perhaps be replaced with "PopStepStack" or something along that line. This is the only occurrence of "_pop".

8.5.6 Data Element Formatting Tags

This was already taken care of by Thinh. 

8.5.7 Data Element Source Tags

Here we also have all lower case identifiers: "camera", "barcode" would become "Camera", "Barcode" and so on.

8.5.9 Data Values

DataMatrixRowValue property Id: Replace reserved prefix "ui-" with "UI-".

8.5.10 Data Variables

System-defined variables (at the end). Replace "system.currentTime" with "System.CurrentTime".

8.9 - 8.11 and globally

Repository identifiers. Already taken care of by Thinh.

BR, Johannes

2012/2/14 Johannes Lehtinen <johannes.lehtinen@rossum.fi>

Hello!

Below is a proposal by Thinh and me for a unified approach to divide identifier value space between predefined standard values and implementation-specific custom values, where both are valid (e.g. repository identifiers, Status Indicators, Formatting Tags, etc).

BR, Johannes

---------- Forwarded message ----------
From: Nguyenphu, Thinh (NSN - US/Irving) <thinh.nguyenphu@nsn.com>
Date: 2012/2/14
Subject: RE: FFMII: Identifier value space
To: ext Johannes Lehtinen <johannes.lehtinen@rossum.fi>

Hi Johannes,

 

Your proposal is basically remove the “system” prefix from all of repositories and standardized is with a unique name starting with first letter uppercase.  As of now, it will impact 3 repositories (system.workTypes, system.users, and system.fir).  I am ok with it to change to :

 

system.workTypes = WorkTypes

system.users = Users

system.fir = FIR

 

For identifier value space, I also agreed with you to use “X-“.

 

So, the global changes are the following:

 

1)      Change all of repositories names to the following:

 

system.workTypes = WorkTypes

system.users = Users

system.fir = FIR

 

2)      Global change all of “x-“ to “X-“

3)      Global change all of the formatting tags in table 58 and 58 tags values from all lower case to first letter uppercase, such as phonenumber to Phonenumber.

 

You can forward my email or your orginal email to the mailing for tomorrow discussion.

 

Thinh

 

 

From: ext Johannes Lehtinen [mailto:johannes.lehtinen@rossum.fi]
Sent: Tuesday, February 14, 2012 6:27 AM
To: Nguyenphu, Thinh (NSN - US/Irving)
Subject: FFMII: Identifier value space

 

Hi, Thinh!

 

Regarding the following open action point...

 

2012-02-08-1: Assigned to Johannes and Thinh: Unify the way to divide identifier value space between predefined and implementation-specific values.

 

Related observation: For some predefined identifier values we use upper case initial "Name" (e.g. Status Categories, Status Indicators, Field-Initiated Request types) while for others we use all lower case "name" (e.g. Formatting Tags, Reference Data repository identifiers). Perhaps we should also unify these to use upper case initials like we have done for most identifiers such as property and type names etc?

 

 

I was considering two alternative unified approaches based on the current spec.

 

We could reserve a specific prefix for standard predefined identifier values (like we have used "system." prefix for repository identifiers) or we could demand that any implementation-specific identifier values use a specific prefix (like we have used "x-" for Status Indicators, Formatting Tags, Data Element source tags, user roles, etc).

 

Considering that we also have static enumerated identifier value spaces that only accept a set of predefined values (such as Status Categories, Field-Initiated Request types, etc) I would prefer a unified solution that does not require any additional prefixes for such static enumerated value sets.

 

 

Therefore, I propose that we adopt an approach where, for any identifier value space that supports both predefined (standardized) values as well as implementation-specific custom values, we would require that any implementation-specific values begin with a particular prefix.

 

Currently the prefix we have used is "x-" (short for eXtension). It could also be something like "ext-" or "custom.". However, I think "X-" (if we decide to go for initial upper case) would be short and fine, and it is also used for the same purpose in some RFCs such as MIME.

 

We could also unify the upper case initial at the same time.

 

 

For example, standard repository identifiers would become:

 

  system.workTypes => WorkTypes

  system.users => Users

 

Some possible custom repository ids:

 

  X-RossumMegacorpProducts

  X-my-fine-attachments

  X-another.repository

 

 

Correspondingly, Formatting Tags would become:

 

  phonenumber => Phonenumber

  monospace => Monospace

 

And custom Formatting Tags would be:

 

  X-Red

  X-Bold

 

 

What do you think?

 

BR, Johannes

 

--
Johannes Lehtinen
gsm +358 40 734 7049
Rossum Oy

--
Johannes Lehtinen
gsm +358 40 734 7049
Rossum Oy

--
Johannes Lehtinen
gsm +358 40 734 7049
Rossum Oy