Where should the first I-joist be set parallel to the rim board

Chis is right in what he is saying. To further elaborate, I assume the structure has a full raftered roof. You need to be absolutely sure if any portion of the wall you want to remove is a supporting part. It is rare that a partition wall running parallel with the ceiling joists is a supporting wall, unless it spans across the center axis of the house perpendicular to the ridge board and some type of post or truss comes from a rafter to the top of that wall. When you look in the attic, are there any angle trusses attaching to the joists, most likely in the center of the building? Is there a center wall on the first floor that runs the length of the house? This is usually indicates a bearing wall. If your wall in question runs parallel with the rafters and joists it most likely is not supporting, especially if there is no joist above it. Since you will have to repair the ceiling after you remove the wall, you will need something to attach your sheet rock to, unless the existing ceiling runs the length and the upper plates of the wall you are removing are located below this ceiling. In any case, you will want to install a joist. If the removed wall is not supporting, it really doesn't matter if you install before or after. other considerations are if there is any electrical or plumbing in the wall. you gotta plan for that as well.

As Tester101 mentioned, in this application there is close to no load on the rim joist, since your actual joists are supported by the beam. Nailing through the rim joist into the end of the other joists will be fine. I've built plenty of decks like this.

Here's a study that was done on end nailing, and the relevant recommendations:

Recommendations

Nails driven into the end grain of wood do resist both static and impact loads. However, additional research is required in several areas. Of the 4,723 data points on the relationship of specific gravity to nail withdrawal strength, only 294 data points were from tests of nails with a diameter larger than 2.52 mm (0.099 in.). Therefore, research is needed on endgrain withdrawal capacity of nails with a diameter greater than 4.1 mm (0.163 mm) so that Equation (5) can be further developed.

Furthermore, none of the studies evaluated repetitive loading of nails from the end grain since joints in service typically do not experience strictly static or impact loading. Repetitive loading may cause end-grain withdrawal capacity to significantly decrease over time.

There is conflicting data on how threaded nails perform in comparison with smooth shank nails when withdrawn from the end grain. Three studies showed an increase in nail withdrawal capacity whereas one study showed a decrease. Since the use of threaded fasteners in construction is increasing, more research is needed to determine the effects of surface characteristics on end-grain withdrawal.

Finally, more research is needed to determine the long-term withdrawal strength of nails driven into end and side grain. The data in this report suggest that over time end-grain and side-grain withdrawal capacity may become equivalent. Very few test replicates measuring the effect of time delay on withdrawal strength were conducted, and more tests need to be run to clarify these limited observations.

If you still feel like you need more support, yes, you can find something like this:

at most hardware stores. If you're working with pressure treated lumber remember you will need galvanized plates and connectors.