Subject: Reference implementations for XLIFF 2.0
I would like to re-start the discussion of reference implementation requirements for XLIFF 2.0. This topic was discussed at the f2f in London, but I don’t believe we reached any final decision/conclusion.
As I recall, there was some debate about the official requirements for reference implementation – it appeared unclear whether we must have reference implementation for everything in the 2.0 standard (each and every module). While that approach would provide optimal assurance and confidence in the standard, it would be difficult to find commitment for extensive implementation work from likely implementers in the near future.
We discussed the various options, which included (to the best of my memory):
1. Only publish modules in 2.0 that have been proven and implemented; anything not implemented would be cut
2. Ask 3 implementers to supply simple proofs of their implementation of XLIFF 2.0 (loosely defined) – i.e. meet the minimum requirements for OASIS
3. Wait until all modules have been provable implemented and delay XLIFF 2.0 publication until such time
There may be other options – please share if there are. However, of the above, option #2 seems the most pragmatic and realistic to me, given the desire to finalize and publish XLIFF 2.0 this year. Option 1 risks publishing a piece-meal standard, while Option 3 risks delaying XLIFF 2.0 significantly. For option 2, there is a risk that some implementation detail in XLIFF 2.0 will not be caught prior to publication, but I believe that is better addressed in a future XLIFF 2.1 version.
Do we have confirmation of the OASIS requirements for reference implementations, with specific detail? If so, can we take a decision on the approach to achieving the appropriate level of reference implementation to meet the timeline that Bryan has created?