Reg, your idea is sound in principle but if you work out the math, it won't fly too well. Here's my back-of-the-envelope assessment.

According to Wikipedia, flow from the well has been estimated to be between 5,000 and 100,000 barrels per day. It turns out this doesn't matter much, as we'll see. The important Wikipedia statistic is the area of contaminated ocean surface, estimated at 6,500 square kilometers as of May 3, two weeks after the spill. That works out to about 465 square kilometers of ocean surface in need of treatment per day. The average square meter would carry about only 34.5 ml of oil, incidentally, representing dilution to a thick film. What with waves and all, we need to suck in a substantial thickness of water to catch the oil over all that area; we're in a hurry here and it's not like skimming a film off a still pond. Let's assume we need to collect and treat an average 0.5 meters depth, about a foot and a half, which sounds on the low side of reasonable to me. That works out to about 6.13E+010 or 60,000,000,000 gallons of water per day.

What's a reasonable pump-and-treat system size? Just as a SWAG, I'll use pumps and membranes for 10,000 gallons per minute (gpm), which is a healthy industrial sized plant. Anything much bigger is going to need a really big boat to carry it. For comparison, you can fit a 500 gpm treatment system into a shipping container, but you'd still need the collection pumps. 10,000 gpm is equivalent to a fair sized creek, big enough for good trout fishing. You could ford it if it's not flowing too fast, somewhere between knee and waist deep, 20 feet or so wide. If you had these 10,000 gpm systems ready to go, you'd need about 4,260 of them, working perfectly efficiently 24/7 with no down time. In practice, you'd probably want at least twice that number both for down time and to cover for imperfect collection. That's a lot of very expensive small tankers hanging out forever, just in case there's a spill in the area. You don't have time to build them; we're looking at probably months to years of membrane manufacturing capacity and really a lot of stainless steel pumps suitable for saltwater duty, not to mention the vessels and trained crews.

You can fiddle with the math if you like but I think my assumptions are pretty reasonable, on the conservative side even. I work all day doing calculations not too different from these. In my professional estimation, you've got a nice concept but the numbers don't crunch.