You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
it will benefit COACH with respect to flexibility, and implement improved separation of concerns, to migrate the text blocks of predefined recommendation text from the CQL into the COACH database, and to replace those text blocks in the CQL with keys that can be passed instead.
this will simplify the CQL, and make it so that if we want to modify recommendation text, all we will need to do is update a database record (easy) as opposed to modifying CQL and rebuilding and repopulating those resources into CQF-ruler (hard). it will also make possible the ability to split what the recommendation text looks like, how it reads, based on other factors (e.g. A-B testing), without introducing more complexity into the CQL and recommendation engine, which is already plenty complex as it is.
this shouldn't be a particularly difficult thing to do either. just replace each text block with a key of some predefined format, separate them by a delimiter, and send that delimited set of keys back to COACH. COACH can then parse that set of keys and build the recommendation text from associated records in the COACH database. it's really a basic search and replace with some DB overhead.
The text was updated successfully, but these errors were encountered:
it will benefit COACH with respect to flexibility, and implement improved separation of concerns, to migrate the text blocks of predefined recommendation text from the CQL into the COACH database, and to replace those text blocks in the CQL with keys that can be passed instead.
this will simplify the CQL, and make it so that if we want to modify recommendation text, all we will need to do is update a database record (easy) as opposed to modifying CQL and rebuilding and repopulating those resources into CQF-ruler (hard). it will also make possible the ability to split what the recommendation text looks like, how it reads, based on other factors (e.g. A-B testing), without introducing more complexity into the CQL and recommendation engine, which is already plenty complex as it is.
this shouldn't be a particularly difficult thing to do either. just replace each text block with a key of some predefined format, separate them by a delimiter, and send that delimited set of keys back to COACH. COACH can then parse that set of keys and build the recommendation text from associated records in the COACH database. it's really a basic search and replace with some DB overhead.
The text was updated successfully, but these errors were encountered: