So you're of the opinion that if something doesn't meet up with claimed performance gains, obviously the purchaser is at fault and it's not possible that the manufacturer overstated claims at all? Or worse, you're saying that if you actually believed a vendor, you're an idiot? If you aren't seeing the benefits claimed, you should just what, shut up and not say anything? I don't agree with that. I don't think you do either and I'm probably oversimplifying your argument, but that's how it breaks down when I look at it. I think if you're in a position to throw some weight around as a company that size is, you go ahead and speak up and call a spade a spade. It may come down to something Facebook is doing wrong, in which case that guy will be eating a lot of crow, but if in fact, there's some overstating going on, and we know there often is, it's in the best interest of everyone if people step up and call them on it. I'll cite the AMD example again. We all know batter claims are BS, so why should we continue to put up with it? The answer is, we shouldn't, but what you or I say on a comment board, even slashdot, isn't really going to make a difference to your hp's and lenovo's and dell's, because none of them are willing to make a realistic adjustment to their numbers unless they are sure the others will follow, otherwise hey they are suddenly getting a lot less run time than everyone else. It takes a company such as AMD to put it out there on the table to get attention.
I think he has some very valid arguments there. Again, specifics and numbers would really help that article, but take a look at google. They basically have the same opinion which is why they build their own units. There's a lot of debate as to if that's a great idea to do or not, but I think google won that debate because they do it and it works. Perhaps Facebook is going to need to go a similar route if server vendors aren't going to deliver what they want. We're all sitting here nitpicking the words of a company that actually runs a huge operation whereas I'd be willing to be bet most of the people here replying (myself included) have never operated anything on a scale that they are talking about. I think we're going to have to wait and see, but I'm hoping there's some follow up articles that detail this out more as I'd like to see the ultimate outcome.