Seagates are very noisy and their performance as desktop drives are not as good as other manufacturers. So, I'd say go for the Western Digital. You may also wish to consider Hitachi drives.

For the external enclosure, yes, you need an internal SATA interface with an external USB2.0 interface. Take a look at Akasa's Integral range, some support both PATA and SATA drives and have an additional eSATA output.

>>2
Seconding the Western Digital recommendation

Whatever you do, avoid Maxtor drives

Incidentally, I have a 320 GB Seagate drive and it doesn't make any noticeable noise above the other things in my case but then again drive noise isn't a priority to me

Never had a problem with Maxtor since the last 10 disks (also, Maxtor is now Seagate but they probably still have separate production lines).

Two of my last 5 Western Digital disks blew up but for their defense, their RMA procedure works well.

One has to be lucky I guess

Google's report on their hard drive failures pretty much ended this debate; all manufacturers were practically identical. It doesn't matter what manufacturer you choose.

Thanks Anonymous for the Google hint, which led to: http://www.theregister.co.uk/2007/06/25/google_barroso_datacenter/
and hence http://www.barroso.org/

But note that the paper does not say that all the manufactuers were identical, indeed it says the contrary. It just says that the paper's conclusions are not changed whether you group by disk series or not:

"Failure rates are known to be highly correlated with drive models, manufacturers and vintages [18]. Our results do not contradict this fact. For example, Figure 2 changes significantly when we normalize failure rates per each drive model. Most age-related results are impacted by drive vintages. However, in this paper, we do not show a breakdown of drives per manufacturer, model, or vintage due to the proprietary nature of these data. Interestingly, this does not change our conclusions. In contrast to age-related results, we note that all results shown in the rest of the paper are not affected significantly
by the population mix. None of our SMART data
results change significantly when normalized by drive model. The only exception is seek error rate, which is dependent on one specific drive manufacturer, as we discuss in section 3.5.5."

My Maxtor survived this! A Miracle!