The GFS model data I've been looking at the last few days has been quite consistent in showing "full" development of Gustav from this point forward, with a minimal cat 3 storm impacting the region from New Orleans to Panama City as early as Labor Day. To narrow it down any further this far out seems pointless. It bears repeating that the biggest difficulty and degrees of error come with intensity forecasts and structure of the storms as they impact coastlines. One things for sure, people cannot say they didn't have enough warning with this storm!