← Prev in month ← Prev in thread

IDs - unit id unique per file (D - 109)

From
Yves Savourel <>
Date
2013-10-22T17:47:54+00:00
ID
006701cecf4e$d3865d00$7a931700$@com
Thread
IDs - unit id unique per file (D - 109)
Hi David, all,

> *D)*
> While it is quite clear that inline ids uniqueness scope needs to be 
> <unit> and not <segment>. There is a similar issue on the <group>/<file> 
> level but the pros and cons do not seem to be that clear.
> Yves requested that unit ids are unique within a <file> element rather 
> than within the parent element, which obviously can be a <group>.
> *Proposed solution:*
> Do not change this, as unit id uniqueness within the parent element 
> seems sufficient and in line with the design principle of requiring 
> the smallest necessary uniqueness scope. Again, this is streaming friendly 
> and makes groups full fledged grouping/structuring mechanism as intended.

I can tell you already that, as the submitter of 109, such "solution" would not resolve my comment.

I'm not following your arguments:

- I don't know where you get the idea that the TC has a "design principle of requiring the smallest necessary uniqueness scope". ->
Not a valid argument for keeping things as they are.

- Uniqueness of unit id within the parent is no more streaming-friendly than uniqueness within a file, as a group can span many
units and even the whole file. -> Not a valid argument for keeping things as they are.

- If the unit ids are unique within each <file> they are also unique within each group. So I don't see how that would impact
negatively making "groups full fledged grouping/structuring mechanism" (whatever that means).

In addition:

- Since your solution allows duplicates across groups, there has to be some benefits to that vs. not allowing duplicates. What are
those benefits?

- I don't think "id uniqueness within the parent element seems sufficient" is fine: Comment 109 gives an example showing that it is
not sufficient, and I guess I'll add a few more arguments:

a) In general IDs should be as unique as possible. XLIFF does not do this in some cases for good reasons. But the burden should be
on demonstrating that having no-uniqueness on a given scope is necessary, not the other way around.

b) Units are one of the basic objects in any localization tool. The capability to identify uniquely each one within a file in XLIFF
would make so many things simpler: e.g. pointing to the unit that has an error, using the unit id as index, etc.

-yves
← Prev in month ← Prev in thread