CRUDdy DBIC question

[Bob MacCallum]

Thanks everyone for the replies - for some reason I only saw them today.

On Mon, Jan 23, 2012 at 5:41 PM, Chris Jack <chris_jack at msn.com> wrote:
>
> You could use something like DBM and there's a section in the Perl Cookbook on using tie with objects but...
>
> I really question the desirability of doing something like this with anything that doesn't pass the ACID test. There are so many advantages to using a relational database (mySQL is free), I'm wondering why you're not going down that route.
>

No, sorry for the ambiguity, we ARE using a RDBMS - postgres in fact.
I meant that we'll load data in from files all the time - new data and
updates (which can include deletes of child objects).  The reason
being mainly that there are pre-existing (bioinformatics) GUI tools
for creating and maintaining the files which take care of a lot things
(mainly ontology term search/suggestion) for us.

> How much data are we talking about?
> Do you care about maintaining your data if your program terminates abnormally?
> Do you need more than one program to access it at a time?
>
> Regards
> Chris