I'm not quite clear what general conclusion the author was going for here, but I take it that one thing he wants to convey is that it would be irrational for a person to consult this data when making purchasing decisions for desktop drives. I don't think he's quite made the point. For one, how does the fact that Seagate admitted to their being a problem with one of their drives make the failure rate of that drive irrelevant? If a car company made a model that tended to fall apart or malfunction due to a systemic problem with one of its systems, this seems like a relevant (though hardly conclusive) reason to think twice about buying a car from that manufacturer. Second, just because the ST31500341AS was first released in 2006, it doesn't follow that the drives Backblaze has are that old. Their average age is only 3.8 years. As you can see from the data (from the chart in the actual Backblaze post, not the one produced in this article), WD greens with an average of 4.4 years fared much better (though the sample is smaller). In fact, they fared better than every other Seagate drive in that table, even the younger ones. I take the point about the amount of IO, but wouldn't it be kind of surprising if these storage pods didn't evenly distribute data? Not doing so would be rather silly. While the specifications say that these Seagates shouldn't be used in high vibration environments, the same is true of the WD drives that performed better. That they handle vibration better seems to me a good indication of long-term reliability in less harsh environments, given that heat and vibration are the killers even there, despite thie not being as extreme. Again, if a car from one manufacturer fares better in tests that push it beyond its intended limits than those from another manufacturer, this seems like relevant information for a consumer looking to buy a car and keep it long-term. I think the author's washing machine analogy speaks to this point. If a laundromat published this kind of data on the failure rate of consumer-grade laundry machines in its laundromats, it seems to me that the fact that one did better than another is a good (though perhaps not decisive) reason for thinking that one of the kind that fared better will last longer in my home than the other. Given the lack of other data out there on failure rates, this is the best information we consumers have, though more information of the type the author is asking for would help to make our future purchasing decisions more informed. What the author needs to show that this limited data is so limited that it is as good as no data, but I don't see that he's done that.

I know I"m off topic here but I feel I must reply. Ethics does begin with considered convictions, but it isn't true that everything is up for grabs (not normatively anyways). Kant's Categorical Imperative does not denote an ethical truth but rather a test for rationality which states that a moral principle must be universalizable in order to be justifiable. For instance, lying is immoral because it fails this test in that one could not wish that the permissibility of lying be universal law. If lying were morally permissible, we could never count on anyone. It is therefore irrational to accept the permissibility of lying as a moral principle. Obviously slavery would fail this test as well.

Further, slavery fails epistemic constraints and is, derivatively, false (in the factive sense). Slavery rests on an assumption along the following lines: Certains races, or otherwise distinguishable categories of persons, are inherently inferior to others and are somehow "built for" a life of slavery. This is as false as the claim that the sun revolves around the earth. I take the burden of proof to rest with anyone who feels that an acceptable moral principle can rest on manifest falsehoods, regardless of a general consensus to the contrary.

The reason slavery is unacceptable today is because the vast majority of the world believes that it is wrong. The ideas of personal freedom and free-will have spread around the world. That is why is it unlawful.

I submit that one of the reasons slavery is unacceptable is because it rests on the above falsehood.

The ideas of personal freedom and free-will have spread around the world. That is why is it unlawful.

lawful != morally acceptable

Descriptively speaking, what is *considered* morally permissible is very much dictated by a general consensus, but normatively speaking (speaking in terms of what ought to be the case), morality is subject to a host of formal constraints.