I feel a strong sense of deja vu... I know we all like XSI on this list but I always feel in these discussions that the the perceived benefit of XSI over Maya is greatly exaggerated due to personal preference. Yes, more marketing of the product is always great but the reality is that the people who make decisions about pipelines are for the most part already familiar with XSI, many have actually worked with it in recent past. It is simply not true to assume that the only reason that studios don't switch is because they somehow are oblivious to its awesomeness.
Unfortunately as it stands XSI simply does not deliver any overwhelming benefits in production over the existing tool set, therefore on a cost/benefit analysis it simply isn't worth the switch. The argument that simply by switching to XSI a studio can ditch hundreds or proprietary tools that were written for Maya is bogus, because most of those tools were written to extend the capabilities of the application and would need to be written for XSI as well because they surpass its capabilities too. Whether or not XSI could be extended in the same way would be up for debate as well! At the end of the day Autodesk needs to make decisions based on the current situation and the foreseeable trends and wide adoption of XSI by mid to large scale studios is just not in the cards. (I'm using Maya as an example, in my current situation I could make the same arguments against switching from proprietary software that leaves a lot to be desired) -- Technical Director @ DreamWorks Animation
