Re: 1984
We all knew they were slimy bribe taking, nest feathering, lying bastards, but I never thought to see the day when they advertised that fact!
I read that as a gentle observation regarding Trump and his new cabinet, but I could be wrong...
4578 publicly visible posts • joined 8 Jul 2009
It helps to train your model on large datasets containing real conversations such as online forums or movie scripts.
Few of those could be described as real conversations. What they've done is chosen texts they can scrape, where someone else has already done the effort-intensive work of typing for them.
Or do you pay Johnny Foreigner to deal with all of that and just trust that they aren't going to do anything devious with your data after it pops out the other end of the VPN tunnel?
I expect most techies would know better than to use a free or cheap, unconfirmed service for anything but casual government-unapproved surfing, in places where you would never ever consider handing over your credit card details or using anything other than a throwaway account. Whether that holds for the non-technical or for teenagers who think they're technical is another matter.
How many participants would be needed, and how long would the duration have to be, for such a study to have sufficient statistical power to isolate all the confounders? Black tea, green tea, tea with milk, tea with sugar, decaff tea, decaff coffee, coffee with muck, coffee without muck, etc, all the levels of consumption, all the possible health outcomes in age-related conditions, complicated by all the changes in consumption and preference that many people go through over time. And, most importantly, what is the likely effect size? How many people gain an extra year (or maybe a number of years) of life, or at least of healthy life, by drinking a certain amount of tea/coffee and by how much would habits have to change for the new knowledge to be worth it? Would the money spent on the study be better spent on encouraging people to do what is already known to make a big difference, like maintaining a healthy weight?
Quite so. My first thought was that this bright star in the sky heralds the coming of the Anti-Trump.
Nostradamus tells us that she will be born as the daughter of the Russian ambassador to the United States. In her infancy she will be guarded by a coterie of kittens and the Groper-in-Chief will attempt to slay her by withholding vaccines from the multitude of deplorables, whose children will die in their thousands.
Continue until Moon orbit corresponds to Earth rotation.
Which is estimated to happen in about fifty billion years time, IIRC.
It's only an estimate because the expanding sun will boil away the earth's oceans long before then, affecting the tidal dynamics between the earth and the moon. And when the sun does expand to consume the inner planets in about 4 billions years' time, the increase in friction will in any case cause the earth and moon to spiral into the heart of the sun and die in fiery dissolution.
But the uni's have stopped caring. So long as you're paying for the year, who cares if you pass or fail or take up the lecturer's time.
Utter rubbish. Student retention is a significant issue and is addressed seriously by universities. Failure reflects badly in the various league tables and in the KIS data, affecting recruitment. No-one is so short-sighted as to think that all that matters is one year's worth of tuition fee income.
Wherever it was you went, they didn't do much to help you improve your critical thinking skills.
Whilst it will be shouted down mercilessly, I seriously suggest that corporal punishment would be an excellent punishment for this type of offence.
There are reasons why corporal punishment has been abandoned in much of the world, and why we decry flogging in Saudi Arabia and other countries. Let's not take a backwards step just to satisfy one person's need to observe another person's suffering. Yours is a gut response to a wider complex problem, as your words suggest you already know.
Not all punishments are going to work as a deterrent with everyone. Removing people from society for years or decades isn't always appropriate either -- it's always expensive and often unjust (look at what the 'three strikes and you're out' policy has done in the US). The justice system in many countries needs reform, and a concentrated effort made on rehabilitation as being the most reliable and cost-effective way of protecting society. That's where all the evidence points.
The answer should be useful whether it works or not.
True, though perhaps only trivially so. It then becomes a matter of how much value you assign to the probability of success compared to the value of failure, and whether you're willing to invest that amount of money into a space-borne experiment.