RE: [all] XLIFF round-trip samples

And here is a bit more well-formed and valid version :)



From: Phil Ritchie [] 
Sent: Tuesday, February 19, 2013 7:51 AM
To: Dave Lewis
Cc: Multilingual Web LT Public List
Subject: Re: [all] XLIFF round-trip samples



An LQI sample attached. 


From:        Dave Lewis <> 
To:        Multilingual Web LT Public List <>, 
Date:        18/02/2013 18:28 
Subject:        [all] XLIFF round-trip samples 


Hi all,
Leroy has placed a series of ITS+XLIFF examples up at:

We've handcrafted these to show incremental annotation of an xliff document by different activities, to form a linked set of transforms as detailed below.  We have a stub for an LQI step, but we've not populated this yet.

Comments on these would be very welcome from: 

* XLIFF-ITS mapping folk (David, Yves): though this not in most cases showing a mapping, it provides some best practice on generating ITs markup within an XLIFF workflow. Later we will implment the ability to generate a snapshot HTML for each stage - to the degree that makes sense - to validate ITS+XLIFF to ITS+HTML mapping. 
* David, Sean will this work with SOLAS OK? In some cases,  you may need to map this and populate the provenance data yourselves. Specifically, could you consume and produce this XLIFF in SOLAS for the MT mapper component if calling Bing (which you won't have to do for Matrex if Ankit can implement the XLIFF interface) 
* Tadej: do the XLIFF input and output for the text analytics step make sense? Would that be a format you'd be willing to support in Enrycher? 
* Marcis: similar question around terminology management (though for the demo we will build a simple front end to demo in Rome - so its a feasibility question, rather than any implementation request) 
* Ankit: we've already discussed using this XLIFF format with Matrex - so please confirm if this is doable 
* David, Sean: 
* anyone interested in mapping to W3C PROV model - we are starting to build a parallel best practice example based on these files at:

the activities map to files as inputs/outputs as follows: 

* extraction: 

* from 
* to

* segmentation 

* from: 
* to:

* text analytics; 

* from: 
* to:

* terminology management (building gloassary from output of text analytics using dbpedia queries for target term) 

* from: 
* to:

* two sets of MT 

* from; 
* to: 
* and then to:

* a PE activity: 

* from: 
* to:

* reassembly 

* from: 
* to:


This email and any files transmitted with it are confidential and
intended solely for the use of the individual or entity to whom they
are addressed. If you have received this email in error please notify
the sender immediately by e-mail.

Received on Tuesday, 19 February 2013 15:13:24 UTC