I love DL and have been using it for years.
It keeps getting better but there is still one *big* design problem.
The way 'buddies' are used in the program is not normalised.
As a Database analyst for many decades this causes many problems in the logic. query and reporting.
What do I mean?
Buddy names are not stored as individuals with 1-to-many relationships. They are stored as unique strings.
so dives with 3 different buddies - in different combinations - are regarded as dives with 9 different buddies
Even if the order of the buddies if different - they become another 'new' buddy string.
eg
John
Peter
Sally
Sally, John
John, Sally
Sally, Peter
Peter, Sally
etc
....are all different 'buddies' (rather than the same buddies, multiple times)
I know this is a schema change and non-trivial but I really beleive its required to resolve the reporting and integrity of the underlying tables. Could you consider a process to normalise the way buddies are recorded in future versions?
Thanks
PS Hope my explantion makes sens but I'm happy to clarify if not.
Normalise the buddies table
Re: Normalise the buddies table
Hi
Thank you for the feedback! I agree, this is not optimal, but changing it won't be easy. I will think about it and discuss with the iOS dive log developer about that.
Kind regards,
Sven
Thank you for the feedback! I agree, this is not optimal, but changing it won't be easy. I will think about it and discuss with the iOS dive log developer about that.
Kind regards,
Sven
Re: Normalise the buddies table
Just thinking about running a report with buddies. Perhaps adding a buddy as "equipment" would be the way to go?
Re: Normalise the buddies table
I'm sorry Irene, I don't understand your question.