My mistake. It had been a while since I went through the process of choosing an API to use with Authorize.net, so I refreshed my memory as to why I will only use DPM:
The advantage of DPM over AIM is this: With DPM, the customer's credit card info never touches my server. The customer fills out the fields on their browser screen, but when they submit the information, it is posted directly to authorize.net's servers. With AIM, the data is posted to my server which then turns around and submits the info to authorize.net.
So while the info is never actually stored in either method, when using AIM the data is at some point present on my server. Which, according to current PCI standards, puts my server is in the PCI envelope.
Also, DPM doesn't require SOAP.
DPM is fairly recent which is probably why there are few plugins and/or systems that have implemented it. So, I am the first to ask for it. Does that mean there is very little demand or does it mean that few people are aware that this far superior method exists? How many people have looked at Hikashop, passed on it because there is no support for DPM and just remained silent (as I did until yesterday)? Can you see how many people have searched your site and forum for "authorize.net dpm"?
Perhaps offering DPM would be a competitive advantage. With all the sample code Authorize.net provides, it doesn't seem like it would be too large of a development effort.
Ok, now that I have tried to make a logical case for it, the next step is just pure begging.