Extensibility

[Yves Savourel <ysavourel@enlaso.com>]

Hi all,

As a first step to continue the discussion on "metadata without extension" I'd like to slay the chimera of "extensibility is bad" once for all.

My understanding is that the community is not opposed to extensibility, it is opposed to bad use of extensions, which is a very different thing.

The real issue is simply that in 1.2 tools are using extensions to overcome the shortcomings of 1.2 and they do it different ways. People complain there are 5 different ways to represent feature XYZ.

The solution is to fix the shortcomings (have one single official way to represent XYZ), not to remove the capability of having extensions. There is nothing wrong with extensions that enhance XLIFF.

In addition I think a large part of the tools vendors are very much in favor of extensibility. That is the only way for them to carry tool-specific data and features. Those are a big part of our community too.


I hope we can, once for all, establish that having extensibility in XLIFF is fine. But: a) they should not be used for duplicating features that already exist in XLIFF. And b) we need to make sure they do not "get in the way".

Now, how to implement extensions is a separate issue.
I'll post a separate email on this since it is a different topic.

Cheers,
-yves