LATEST UPDATE: I finally got a replacement and looked closer at the condensation in the tank. I discovered a lot about the design of this coffee maker that I think is relevant to the original mildew issue. One important thing I found is that it isn't just steam that flows back into the water tank, it's also hot water, and it seems to be intentional. During brewing, or after brewing when the brew basket lid is opened, a little hot water leaks out of the water bridge, and the large open passage between the basket and tank is meant to drain this bit of water back into the tank instead of letting it go into the pot. Thus, not only is steam flowing through and condensing in the tank, but hot water is flowing back in, also. Another important thing I found is that the condensation in the tank also comes from the boiler itself. The condensation forms due to the latent heat in the boiler vaporizing water remaining inside, and this vapor then flows back into the tank and condenses on the wall. I sealed up all the open spaces between basket and tank using foam weatherstripping, as I'd suggested previously. I used the largest piece of foam in the "drain" space beneath the bridge, and because it contacts hot water that can enter the pot I coated it with food-safe silicone sealant. This decreased both the amount of steam and the amount of liquid water flowing back into the water tank. There's still some condensation on the tank wall, as I noted above, that comes from the boiler itself. Will this reduce the risk of mildew? Possibly, because of the lower amount of water condensation. But I think it will help in another important way. I've been trying to figure out where the mildew got its food. It had to have some organic matter to eat in order to grow. There are three possibilities: 1) There's a material of construction in the tank that mildew can eat -- very unlikely, as mildew need material that can be decomposed by their enzymes, and that's probably not going to happen with most commercial plastics. 2) When filling the tank I either touched the inside with my hands or some bit of organic junk fell inside -- very possible. 3) The coffee in the brew basket was the source of organic matter, and a bit of it was transported through the big "drain" hole into the tank due to the flow of steam and liquid hot water -- this is also very possible, and I think even likely, because coffee is a great medium for growing mold and there's a lot of it not far from the hole. Therefore, blocking up the opening should prevent this from happening. So why is the machine designed with this way? Beats me. I can't think of a good functional reason for this design, so I can only assume it was a cost driven choice. Thus, I'm not going to change my rating. I think this is a design oversight, and OXO can do better. However, the machine still makes great coffee, and I emphasize that so far only one other person here has had a mildew complaint, so the odds are there won't be an issue for most. ORIGINAL UPDATE: I have found a significant flaw in this machine. Steam from the brew basket flows back into the water tank, which then condenses to water that doesn't evaporate away because of the closed tank lid. The dampness apparently caused mildew to grow in the tank, probably in the seams and joints where the clear tank meets the rest of the machine. This made the tank smell, and several times the coffee ended up smelling and once even tasting of mildew. Also, I was unable to eliminate the mildew odor because I couldn't immerse the machine to soak it. You can probably avoid this issue by leaving the lid to the water tank open in order to keep it dry, but then you expose the tank to dust and germs. However, I think it can be fixed by using a piece of waterproof foamed weatherstripping to close up the space that joins the brew basket to the tank. You'll have to cut it to fit, and make sure it doesn't touch or interfere wth the water bridge that connects the boiler to the shower head. I'd also recommend coating the foam with some food grade silicone adhesive to keep it from possibly leaching chemicals into the coffee. Of course, you could also just use a big gob of the food grade silicone adhesive to seal up the space. It might be messy, but it should work just fine. Another simple solution might be to just leave the tank lid open between brew cycles but cover the tank with a clean cotton towel. That should allow the tank to dry out without letting foreign matter into it. In fairness, I'll note that so far I seem to be the one "lucky" person who has had this issue. Still, I've never seen a coffee maker that connects the brew basket directly to the water tank, and even the possibility of this mildew problem could have been avoided by not doing this. It's a pretty stupid design flaw, and for now my review goes from 5 stars to 3. If I can successfully fix it as I described above, then maybe I'll raise it to a 4. --------------------------------- I've spent a year looking for a coffee machine that 1) makes proper coffee; 2) is well made; 3) has a decent warranty from a recognized national brand; 4) doesn't cost a small fortune. The OXO works on all counts, but as I mention in my update it has a really stupid design flaw that limits it to 3 stars. What's good: The coffee: Use a proper medium coarse grind and decently good quality beans, and the OXO will deliver. Here's where I'm going to go into detail because I will make comparisons to two other popular pour-over machines I tried, the Bonavita 1900TS, and the Bodum Bistro. The OXO is SCAA certified, and I have tested mine as close to the SCAA conditions as possible to prove that it does actually meet the very important temperature specification set out by the SCAA. With a fine wire thermocouple bead probe (a proper thermocouple with a proper junction compensated readout), I measured the temperature at the top of the grounds during the brew cycle (according to the SCAA procedure). It was 200 F +/- 3 degrees in the wet grounds for most of the brew cycle. This is spot on to the SCAA requirement. The exit temperature of the water from the head was about 207 F. The SCAA also tests at a position near the bottom of the grounds, but I didn't do this. The Bonavita is also SCAA certified, but it did NOT meet the temperature requirement when I tested it. I actually tested two, because I exchanged the first one after determining this problem. Both exceeded the SCAA 205 F limit by the middle of the brew cycle and got as high as 210 F in the grounds, and water reached boiling temperature coming out of the head. The Bonavitas also made bitter coffee, and I believe it was due to this overheating. The Bodum is not SCAA certified, but performed like it was. It peaked at 204 F in the grounds, and it's coffee was much better than the Bonavita, and just a little behind the OXO. This might be because it took longer to get into the right temperature range than the OXO. The 8 cup brew time is about 6.5 minutes for the OXO, 5.5 minutes for the Bonavita, and six minutes for the Bodum. The shower head: The OXO has a six hole shower head that covers a good deal of area, and which wets the grounds very evenly. After brewing, there are divots in the grounds beneath each hole, indicating that flow is uniform out of the heads. The Bonavita had a similar shower head that covered a larger area with more holes, but the uniformity of wetting was not good. As with many others, both of mine dumped water out of one or two holes only. Also, I found that after disassembling and reinstalling the shower head of the Bonavita according the the cleaning instructions, the head would leak along one of the assembly seams. The Bodum's shower head was fantastic. It was all silicone, and had many holes over a wide area. It saturated the grounds perfectly, and there were so many holes that the grounds looked completely wet and even, without any divots beneath the holes. The carafe: I tested