Follow Slashdot blog updates by subscribing to our blog RSS feed

 



Forgot your password?
typodupeerror
×

Comment Re:Evidence based slashdot comment (Score 1) 95

And what are the chances that a drone travels across the galaxy using who knows what sort of beyond magic tech yet at the last minute fucks up its re-entry and burns up?

Riiiight.

In all seriousness, a modest sized drone is actually something that makes some sense to crash.

I can easily imagine us earthlings building 10,000 drones and flinging them out into the stars on a voyage of discovery knowing full well that 99% will malfunction in various random ways over their many century journey and not return any data. One of the likely "not return data" states could be described by "I actually got to the intended planet, but I my computer is too damaged by cosmic rays to find a stable orbit".

Comment Re:Quora Has Been Useless for a Good While Anyway (Score 1) 57

Ack. Well, your comment has convinced me that I should simply stay away from Quora. I took a recent break because it was annoying me with too many crap questions. Now I know that it is not never going to get better; the algorithms are going to favor a short term increase in activity over long term value (as I perceive value).

Comment Re:Grasping at straws (Score 1) 77

Yes, drug development is a very specific and very capital intensive kind of business. It may look good at the slide deck level to wax poetic about how all that genetic information is going to help you pick the winners. But the reality? Not clear whether it actually helps, even a little.

Also, in most cases, your drug has to work pretty well in the first place, and then you can imagine how genetic information makes it more effective in finding the right patients. But that is a difficult and expensive follow on research project.

My take on this is Wojcicki found herself sitting on a big pile of cheap cash, and went with drug development because it sounded like a good way to make big money because she lacked original ideas of her own. I am not saying that was a definitely bad path. But how exactly is it a good idea for 23andme to play the game?

If I want to invest in drug development, I can invest in startup companies that do that or funds that invest in those startup companies. Sophisticated investors are not going to be impressed with a company of slapped together businesses, even if the pieces slapped together are okayish individually. Now, you have a big pile of diversified stuff, Wojcicki. Lovely. Is your stuff profitable?

Comment Re:Good (Score 1) 32

It boils down to a sensitivity vs specificity question.

The idea of simply arresting someone based on nothing more than a witness lineup proves extreme laziness. Witness identifications are wrong all the time, and throw enough somewhat similar to the description pictures in front of the witness and they are likely to pick someone eventually. With the right wheedling, I am sure a detective could find someone to arrest 99% of the time. While that is good sensitivity, we do not need police officers who cannot be bothered to think about specificity. There are likely to be hundreds of people in the city who look somewhat similar to the perpetrator, and arresting a random one of those hundreds is the kind of police "work" that erodes trust.

Comment That is a very reasonable question (Score 1) 138

The rate of heat exchange off the board is proportional to the amount of "stuff" that comes in close contact. Liquid is "more stuff" than gas. Mineral oil is ballpark 100X as dense per cubic centimeter as air.

What that means you can pull 100X times as much heat away from the board with the same volume of material flowing by. Now, such things are not equal, because it is sure easier to blast air through a server than liquid.

But I do not think these two stages are where the big win comes from, because they somewhat offset. The win is at a third stage where the "more stuff" argument also applies to when you try to dump the heat outside into the environment. You can design very efficient heat radiators that cool this hot mineral oil.

Keep in mind, that data centers are already past the point of just using normal "room temperature" air. They must take outside air, cool it down with air conditioners to get it to frigid temperatures, and then blast that air into the racks, and finally throw the now very hot air outside. With mineral oil as a heat exchange medium, I do not need expensive air conditioning. Simple pumps that pump the oil and the oil is cooled in radiators that have simple fans blowing air across them. A fan and a simple pump takes a lot less electricity than any refrigeration type system like a powerful air conditioner.

Comment Re:and the beatings will continue until moral impr (Score 2) 49

That strongly suggests the managers should improve their oversight skills. Perhaps certain individual contributors should be let go, but that is the symptom, not the root problem.

These large job cuts in very profitable companies indicate the top level decisionmakers are making big decisions based on emotions. They overhired in the past because of emotions, and now they decided to do something drastic based on their emotions about perceptions of the stock market. Two wrongs might well be right, but I do not think "wow, these tough leaders are doing the right thing" but "these clowns could do this right, but they are doing something quickly for lack of courage"

Slashdot Top Deals

He who has but four and spends five has no need for a wallet.

Working...