The example Thiel used in the original "seriously but not literally" argument was the wall. Of course Trump wasn't talking about a literal wall, he said, it was more of a metaphor for, you know, something.
This argument was always embarrassingly stupid, but anyone still making it after Trump shut down the government for a month over literal wall construction has forfeited their right to be taken seriously at all.
Sadly, you'll still find people (even in this thread) that claim "of course Trump's campaign wasn't meant to be taken literally." With statements like "if you took what Trump said on the campaign literally, then you missed the point."
So then if _The Wall_ was figurative, why did he shut down the government over its funding?
"No he meant that one literally."
Okay... So what about the fact that the wall would be funded by Mexico? He claimed that he would literally make them write a check.
"That was obviously figurative."
An exhausting stretch of logic by those who can't burden themselves with introspection.
Like any politician, the "Mexico would pay for it" was an "argument" to get support. Did supporters care if Mexico itself would pay? No. But it fit their sentiment. It's win-win for him because the only people who would "care" would be people against the wall who politically don't matter.
It's like when Dems say they want to abolish ICE. That's what their base want to hear. Do they really think ICE is going to be abolished? No, do they like hearing that because it rings nicely in their ears. yes!
Let's abolish ICE is the lefts version of "Mexico will pay for the wall". When they say Abolish ICE, they mean we'll (try to) do the dreamer things and we'll be less literal on refugee interpretation (and include economic migrants as definition of a refugee) for example. And, if it doesn't happen (not enough internal support) we'll blame the Repubs, no loss on their side. This is how political comms works.
Don't try and hang them on each and every word. You'll miss what they are saying.
> It's like when Dems say they want to abolish ICE. That's what their base want to hear. Do they really think ICE is going to be abolished?
Yes, they do.
They may not think the functions are going to be eliminated, or most of the rank and file staff removed from government service, any more than when the predecessors of various parts of ICE, like the INS, were abolished before it.
It is intellectually dishonest to attribute literal intent to some of his statements, but others made with the same bombastic style only obviously more ridiculous are hand-waved away as "it was the spirit of the thing he meant, not literally". And whether or not his followers care about Mexico paying for the wall, his various statements clearly indicated a literal intent on the topic. And while some on the left may mean "let's massively reorganize ICE and rethink its operational policies", it's clear that some also mean this literally.
You're right about Occasio-Cortez: she is shaping up to be about as bad as Trump in this respect. I'm not quite sure which of them is worse, but I know the prospect of her someday getting to the White House is a chilling prospect, even though my political leanings are in that direction. She appears to be about as dangerously ignorant (and perhaps callous) of facts of situations. The only mitigating factor for her is that perhaps experience will temper this, and maybe she is just uninformed but not unwilling to learn.
I disagree with Thiel there, His base, and most independents want a literal wall (and, as he claims, Dems used to be for a wall) But his speeches are littered with these blue collar type speech patterns (boasting, exaggeration, put-downs, etc., when trying to “win” an argument).
> His base, and most independents want a literal wall
Can you back up that most independents want a wall? The polling I've seen says otherwise.
"CBS News polling from mid-November found that a majority -- 59 percent of Americans -- oppose building a wall along the U.S.-Mexico border. It's a partisan issue, though. A large majority of Republicans support the wall -- 79 percent. A majority of independents -- 66 percent -- oppose the wall, and 84 percent of Democrats are also against it."
Can we stop saying that "blue collar" people delve in boasting and exaggeration ? Most blue collar people I meet are very polite, mild mannered and focused on executing their tasks. The only people who talk like Trump are narcissists.
> This argument was always embarrassingly stupid, but anyone still making it after Trump shut down the government for a month over literal wall construction has forfeited their right to be taken seriously at all.
Whether his statements during the campaign were taken literally or not it appears he was taken seriously by enough people to win him the nomination and the election. I think more than anything they bought into his intent. And his values however we may choose to the define them resonated with them.
What’s really hilarious is that the wall was originally intended to be figurative. His staff came up with the idea as a way of keeping Trump’s scattered brain focused on talking about immigration. Except the moron ran with it.
They settled on Condorcet, a method almost unheard of outside of electoral/political theory, and your criticism is that they didn't make use of political theory?!
That defines the voting system, but not governance and representation
It's also almost exclusively used by other software foundations like the Debian SF so if anything this just proves my point that programmers by and large are not equipped and do not invest time to become equipped for discussing governance.
Fair. Perhaps the most salient criticism is that Zuckerberg's top priority is connecting people because his company is designed to use those connections in order to sell lots of advertising.
Advertising is the best way to pay for the servers, lawyers, developers, office space, and everything else FB needs, while keeping the cost for users low.
I believe they are investigating options for letting you pay for FB to turn off advertisements, however.
In the end, it's hard to know for sure what Mark's true motivations are, just as it is for anyone else. The only person who knows Mark's true intentions is Mark. That said, I personally believe him. When I listen to him talk about what he wants to do with Facebook, it seems obvious to me that he has good intentions at heart.
He's about my age, and when I was young the internet let me communicate with people from different areas of the world, different backgrounds, and helped me expand my own understanding of humanity. The internet made me a better person -- a more tolerant, more thoughtful, less prejudiced person. I hoped that the internet could facilitate the same personal growth in the rest of humanity, too. By drawing people together from the disparate corners of the world online, we could become a more tolerant, more understanding species. Mark has said similar things.
Unfortunately, things haven't quite worked out as he, or I, hoped, it seems. =(
Is this a trivial change in user behavior from using Chrome?
(My assumption is that, at the very least, I'll need to switch from using my password manager extension to using a separate app, and having to copy/paste. I'll also probably have to export/import bookmarks. I also use Chrome profiles heavily. Are all of these things trivial to migrate?)
The Compiz Cube was cool as heck and no one will ever convince me otherwise. I used to have it configured to layer the windows in 3D space based on how recently they had focus, had it snow and rain on my desktop to match the actual weather using cron, curl and a few compiz plugins, had hotkeys to change transparency or toggle always-on-top or invert colors when I had eyestrain - gods I miss that era.
Like everyone else, I use macOS now because it's what work supports, and the desktop is nearly unusable and primitive as anything. Every day I rage that I can't adjust transparency or toggle always-on-top however I want.
You are so right. I loved that cube, and there were other customizations that truly worked, but we've traded them all for reliability, a little bit of consistency, and other compelling things. But that era felt promising it's still a little bewildering that things didn't continue on that path.
The slogan is "Yes in my backyard." But the people saying it don't have a backyard. They should not be coercing people with backyards and imposing their will.
If I live in SF, I have the right to vote in local elections, and I have the right to vote for people who represent my interests, regardless of whether I rent or own. If you're hoping for the ballot box to be restricted to homeowners, you're about 300 years too late.
It's more conversationally natural for me to refer to the yard adjoining my property as 'my backyard' than my landlord to (who likely has a yard of his own for which that is more semantically assonant).
That's technically wrong, the worst kind of wrong. If you rent a house, the backyard is owned by your landlord. Sure in casual conversation you will say "my backyard", but we're talking about ownership here.
And we're talking about how a (loose) organisation is branded, not the legal implications. The "backyard" in question is already one metaphor deep, for crying out loud.
This is like complaining that Amazon has little to do with the rainforest.
Jesus Christ, it's not literally referring to the backyard attached to your particular house. I can't believe I have to explain this, but "backyard" in this context means neighborhood.
That's a feature, not a bug. High cost is an efficient barrier to growth that exceeds the carrying capacity of the region.
California is suburban. Changing that will take decades. Cramming more people in to the Bay Area will not result in transit solutions springing up...these will take decades.
Don't forget that the State continues to have long term systemic issues providing water to current residents.
California is at 40 million now and I have no desire to see it reach 50 million. The Governor needs to start thinking about how many people we want here.
It's training data. There's 17 suicidal and 17 non-suicidal scans, for a total of 34 scans. They trained 34 models, leaving one scan out each time. Of those 34 models, 31 correctly predicted the left-out scan.
IANAStatistician, but this seems like a trash result.
Cross validation is ok if you do it once, but they repeatedly did it and chose the features based on the results. You can't keep adjusting your model/features based on cross validation performance without overfitting to the training data.
In this case nested cross-validation would have been the proper way to do this. Run your entire model selection process (scaling - feature selection w/ CV - model selection - hyper paramter tuning w/ CV) on each of the folds in the outter CV loop. That will tell you how good your process is at building a model that generalizes.
"The features used by the classifier to characterize a participant consisted of a vector of activation levels for several (discriminating) concepts in a set of (discriminating) brain locations. To determine how many and which concepts were most discriminating between ideators and controls, a reiterative procedure analogous to stepwise regression was used, first finding the single most discriminating concept and then the second most discriminating concept, reiterating until the next step reduced the accuracy. A similar procedure was used to determine the most discriminating locations (clusters)."
The features were chosen using the same data as used to assess predictive skill.
That quote does not support your summary, unless you are basing it on the information not explicitly mentioned. (I.e. they didn't say that they were only using training data to select features, but if they are any competent, they did.)
Select a training set, leaving out one sample for validation. For all features, train a classifier on the training set using that feature. Keep the one that gives the highest discrimination score on the training set. Repeat with more features. Then evaluate the final classifier on the validation sample, which has so far not been seen in any of the steps. The result provides an estimate of the risk on unseen data from the same distribution.
To get the estimation variance down, you can repeat this for all possible choices of validation sample. That means, you start the feature selection process on the new training set over from scratch and obtain another risk estimate. If they kept the features selected earlier, that estimate would be "contaminated" and not independent, but if they correctly start over, the procedure is valid.
My understanding is you are saying create N (N=34 in this case) different parallel models that use different features/etc. Then take the average (or whatever summary stat) of the accuracies to get the predictive skill.
When we want to use these models, we run new/test data through all N=34 models in parallel and calculate a prediction from each. Then somehow these predictions need to be combined (one again an average, etc). This is the average of the predictions, not accuracies/whatever.
Where was the step combining these predictions present during the training? It seems your scheme necessarily calculates an accuracy based on a different process than needs to be applied to new data.
No, when you want to classify a new sample, you take a model trained on the complete labeled data you have and use the prediction of that. The validation procedure using those 34 models trained on subsets of the data is just to tell you how accurate you should expect the result to be. Afterwards, you can throw those models away.
Of course you could build an ensemble model, but if you want to know the expected accuracy of doing that, you need to include the ensemble-building into your validation procedure. (Or use some theorem that lets you estimate the ensemble performance from that of individual models, if that is possible.)
I see. So usually what you would do is run the CV a bunch of times to test various features/hyperparameters, knowing this will overfit to the data used for the cv.
After deciding on features/hyperparameters (based on the overfit cv), you train the model on all the data used for cv at once. Then test the resulting model on a holdout set (that was not used for the cv). The accuracy on that holdout would then be the accuracy to report.
This sounds much like what you are describing, except you only do one cv and do not use it to decide anything. The cv is only to give an estimate of accuracy.
Is that correct? It does seem to legitimately avoid leakage. However, it seems impossible that an anything close to optimal feature generation process or the hyperparameters were known beforehand. Do you just use defaults here?
Nothing is free - someone is paying for it. When you give people specific things instead of money, you're taking choices away from them, insisting that you know what they need better than they do.
If you give people money, they can decide what they need most and go do that. Maybe I don't need housing because I can live with relatives and use my money for education. If you insist on free housing instead, you take that option away from me.
Letting people make mistakes is important. Letting people be altruistic is important.
More specifically: I'm 0% worried about sending money overseas, since I expect that the few people who do so that will be making the world a better place on net.
This argument was always embarrassingly stupid, but anyone still making it after Trump shut down the government for a month over literal wall construction has forfeited their right to be taken seriously at all.