This is the time for QBMS users to weigh in. There are two approaches to how QBMS handles "duplicate transactions", and I would like feedback - if any - on what users here think. The next release, which is imminent, relies on this.
Option 1 - ignore the entire duplicate handling that QBMS does - and treat every single submission by a cart user as "valid" and will be processed. Downside to this, and the reason QBMS created the duplicate algorithm, is to handle the possibility of a double click on submit, or if the response does not come back at all (network issue for example), and the user re-submits. Then you end up with two charges for the same order. So Intuit created the "duplicate submission" algorithm on the server side, but it relies heavily on good coding and transaction ID checking on our side (something which has not been taking place).
Option 2 - use the duplicate handling (as we have) which says that if the SAME customer submits an order for the EXACT SAME amount within 5 minutes - it will be considered a duplicate on QBMS end. Reasonable - but up until now, the order would still be a new order in Zen, but not a charge at QBMS. The new 2.0 code handles this by checking if the transaction ID has been issued before - and if it has, it generates an error at the user end that the transaction appears to be a duplicate (same customer, same amount) and does not process or create the new order. If it's in fact a valid second order, they have to wait 4 more minutes to resubmit.
So... what do you think? Go with the enhanced duplicate handling code for 2.0, or eliminate all duplicate handling at all as part of a 1.9 release? Either way - both versions are written and tested - but I need to get one of them up there promptly as 1.8 has a bug in that code and we need to publish that fix.
Bookmarks