Page tree

April 2022 BETA Version 10 (termMed build via RAD) vs April 2022 Production Version 1 (termMed build via RAD) traceability

Differences found in package Comparison

Number of RF2 records impactedRelated JIRA ticket(s)Rationale
Readme filen/a

x Prefixes removed, + Production naming convention added in as expected.

Also, ALL Delta files removed as requested.

October 2021 PRODUCTION Published (termMed build via RAD) vs April 2022 BETA Version 10 (termMed build via RAD) traceability

Differences found in package Comparison

Number of RF2 records impactedRelated JIRA ticket(s)Rationale
RefsetDescriptor files2

2 new records added - looks like for a new Language Refset (900000000000509007) - the US Language Refset:

> 45f5491a-df0a-b7e1-53cc-f71e93c12ba5 20220430 1 450829007 900000000000456007 900000000000509007 900000000000511003 900000000000461009 1
> 55291b07-5048-392c-c5e2-f019d1d22ee5 20220430 1 450829007 900000000000456007 900000000000509007 900000000000510002 900000000000462002 0

Emailed Guillermo on 26/4/22 to confirm that this was intentional.... they confirmed that it was, via email at 20:00 on  

ModuleDependency files3

3 records added for April 2022 release, as expected?????? (Wasn't it 1 record last time???????)

> 23db3c9c-0ecd-6b43-d93b-99bce861ce65 20220430 1 450829007 900000000000534007 900000000000207008 20220430 20220131
> 242e19f9-8b64-4c6e-9ff7-f84b7bc60be8 20220430 1 450829007 900000000000534007 449080006 20220430 20220131
> 2a3c4d38-43c2-4514-ae22-6cbcb2692847 20220430 1 450829007 900000000000534007 900000000000012004 20220430 20220131

The first record is the normal expected record change:

> 23db3c9c-0ecd-6b43-d93b-99bce861ce65 20220430 1 450829007 900000000000534007 900000000000207008 20220430 20220131

However, the second two are unexpected:

242e19f9-8b64-4c6e-9ff7-f84b7bc60be8 20220430 1 450829007 900000000000534007 449080006 20220430 20220131
2a3c4d38-43c2-4514-ae22-6cbcb2692847 20220430 1 450829007 900000000000534007 900000000000012004 20220430 20220131

Emailed Guillermo on 26/4/22 to confirm that this was intentional:

  • They appear to be calling out dependencies of the Spanish module on both the Component Model module (potentially expected given all the recent discussions on dependencies not being transitive), but also on the ICD-10 module? 
  • Were these changes intentional?  If so, this would be another worthy inclusion in the Release Notes…
    • Guillermo confirmed that they were, via email at 21:26 on  
Readme filen/a

Updated for April 2022 + Beta "x" prefixes as expected

HOWEVER, I noticed the .json file was now included, and wasn't populated correctly, so have asked Ale to fix this...

ALSO NOTICED DELTA FILES STILL IN PLACE - ASKED ALE TO FIX URGENTLY...

Concept files0

0 files added/updated in Release package files

Emailed Guillermo on 26/4/22 to confirm that this was intentional....

- Ale confirmed this was as expected, via email at 20:00 on  

AttributeValue files2947

2746 records added +

197 records inactivated +

4 records updated

Emailed Guillermo on 26/4/22 to confirm that this was intentional....

- Ale confirmed this was as expected, via email at 20:00 on  

Language refset files39637

29174 records added +

9754 records inactivated +

709 records updated 

Emailed Guillermo on 26/4/22 to confirm that this was intentional....

- Ale confirmed this was as expected, via email at 20:00 on  

Description files852233 

28128 records added +

9651 records inactivated +

814454 records updated 

- Guillermo confirmed that this HUGE amount of change was as expected, via email on 26/04/2021 at 18:05, as there were 812662 records in the Delta!   His explanation of why was:

Just a heads up, there are 812662 modified descriptions in this release, a very large delta. We finally went ahead addressing the caseSignificance changes after the English edition policy changes years ago. The Spanish Edition, like many other translations, usually starts with lowercase (uppercase is only used if significant), so before the English edition change, maintenance of initialCapitalStatus was straightforward. Australians introduced in RF2 specs an explicit "entire term case sensitive" to support tall man lettering, but later reinterpretation of the documentation brought "entire term case insensitive", a change made in batch in the English IE. Beyond producing ~800,000 changes with the algorithm, we delayed the implementation because of several concerns. One of them was the identification of terms containing all lowercase letters that were case-sensitive. Our estimation was that we needed to manually review about 18,000 descriptions to evaluate whether we could change them to case insensitive with an enhanced algorithm. In this release we had reviewed about 80,000 descriptions looking for special cases that could escape the algorithm settings, and then went ahead with the changes.

TextDefinition563

535 records added +

28 record inactivated

- Ale confirmed this was as expected, via email at 20:00 on  


  • No labels