Share good ideas and conversation.   Login, Join Us, or Take a Tour!
joelthelion's comments

Could this hurt Trump's campaign?

kleinbl00  ·  785 days ago  ·  link  ·  

We can only hope.

And bring it up in conversation.

Note that as bad as this is, it only affects some statistical methods used to analyze fMRI scans, not fMRI as a whole. fMRI is an amazing tool.

I think once again we are seeing the failure of the "publish or perish" model of academic research.

jadedog  ·  858 days ago  ·  link  ·  

Could you explain this a little more? How does the inaccuracy of some statistical methods used to analyze fMRI scans not affect fMRI as a whole?

Without knowing a lot about it, it would seem like if something affected a subset of something, then the whole thing is affected.

Devac  ·  857 days ago  ·  link  ·  

    Without knowing a lot about it, it would seem like if something affected a subset of something, then the whole thing is affected.

That does not have to be true. The article talks about errors of the I kind (or alpha errors). It's just a fancy name for a false-positive result.

Now, when you are testing multiple hypotheses you have multiple statistical tests done on the gathered data set. Because false-positive can vary by hypothesis tested, you have to account for it. You do that by finding a control level (how strict adherence to hypothesis we want to have) for this error. If not accounted for it, you just did what is called a Family-wise error.

jadedog  ·  857 days ago  ·  link  ·  

OK, I think I might be getting it. You can tell me if this sounds right.

This is the example I was thinking of (from the wiki you linked on alpha errors).

    Perhaps the most widely discussed false positives in medical screening come from the breast cancer screening procedure mammography. The US rate of false positive mammograms is up to 15%, the highest in world. One consequence of the high false positive rate in the US is that, in any 10-year period, half of the American women screened receive a false positive mammogram. False positive mammograms are costly, with over $100 million spent annually in the U.S. on follow-up testing and treatment. They also cause women unneeded anxiety. As a result of the high false positive rate in the US, as many as 90–95% of women who get a positive mammogram do not have the condition. The lowest rate in the world is in the Netherlands, 1%. The lowest rates are generally in Northern Europe where mammography films are read twice and a high threshold for additional testing is set (the high threshold decreases the power of the test)

The false positives are higher in the US than in Northern Europe, not because of the equipment or procedure used, but because of the stringency of the test.

Because of that, false positives might be high for a certain set of data where fMRI technology is used but not for all data where fMRI technology is used.

Devac  ·  857 days ago  ·  link  ·  

It could be that, but there are also procedural factors as in the one where you have cited.

For example, to my knowledge (at least in Poland) when you are tested for HIV or other infections that can linger undetected they take two samples. That way you need a follow-up only if both of them show different results, since this is how you can usually catch false-positive. There is still a fraction of people with both samples being false-positives, but it is order(s) of magnitude less than with one sample. This actually a great example of Bayesian Process, one of the more important concepts in probability and applied sciences.

I can try and write an example, but I think that it will save both of us some time if I'll just link this:

and here is some extra footage that is also relevant:

I can explain it further if you would have follow-up question after video, so don't take it as me brushing you off ;).

Devac  ·  859 days ago  ·  link  ·  

Perhaps you are right, but to my knowledge sloppy stats must be shown, demystified and corrected. Kept as a sort of warning for future reference so to speak. In the end, reason behind publication (the publish or perish that you have mentioned) is less important than the results it brings to the field.

I think the lesson is that a lot can be accomplished without modern languages and practices.

So we shouldn't worry too much if we're not using the latest trendy framework and instead focus on the software we are building to make it great.

HumpbackWhale89  ·  756 days ago  ·  link  ·  
This comment has been deleted.
joelthelion  ·  874 days ago  ·  link  ·    ·  parent  ·  post: The problem with reinforced concrete

They mention a couple potential alternatives in the article.

kleinbl00  ·  874 days ago  ·  link  ·  

Yeah. "mud brick" and "rammed earth."

I repeat: nobody is going to build office parks out of cob.

Sorry, this comment is private.
joelthelion  ·  875 days ago  ·  link  ·    ·  parent  ·  post: The problem with reinforced concrete

I disagree. I don't think rebuilding everything every 50 years is very sustainable, especially considering the very high CO2 emissions when making cement/concrete.

So of course using plain concrete is no solution, but looking for better solutions is very much needed IMHO.

kleinbl00  ·  875 days ago  ·  link  ·  

Name a single building material with better sustainability and durability.

Nonody is going to build office parks out of cobb.

joelthelion  ·  874 days ago  ·  link  ·  

They mention a couple potential alternatives in the article.

kleinbl00  ·  874 days ago  ·  link  ·  

Yeah. "mud brick" and "rammed earth."

I repeat: nobody is going to build office parks out of cob.

bioemerl  ·  875 days ago  ·  link  ·  

    I don't think rebuilding everything every 50 years is very sustainable

Cost reflects sustainability. These green techs are often very cute, but more often than not they drain so many resources that they will ultimately cost us more total pollution than the cheaper often.

In 100 years tech advances so far that rebuiding saves more energy and time in the long run than leaving the building does.

It might be a bit early to say "the dust has settled", but good analysis nevertheless.

For as far as I can remember, the European Union has been governed by dogmatic technocrats with no concern whatsoever for ordinary people. I'm not saying they didn't do a lot of good things, which they did. But at some point you need to look beyond growth numbers and check that ordinary people are actually benefiting from your policies.

joelthelion  ·  877 days ago  ·  link  ·    ·  parent  ·  post: Inside a Potato Gun Explosion at 20,000 fps

Really cool, thanks for sharing!

joelthelion  ·  886 days ago  ·  link  ·    ·  parent  ·  post: Bitcoin's mempool is over 20 MB

I really wonder where bitcoin is going:

- The community is divided over the block size issue

- One of bitcoin's biggest forums, the bitcoin subreddit, exploded due to censorship issues, making community division even worse

- Gavin Andressen, one of the only really trusted bitcoin developers, lost a lot of credibility when he announced Craig Wright was Satoshi

- The halving is due in less than a month

All in all an extremely complex situation, I find it hard to foresee anything at all in this mess. Maybe at least hubski could be a good reddit replacement for discussing bitcoin?

mk  ·  886 days ago  ·  link  ·  

I think it only now has a shot as digital gold. That might just happen, but I think it's unlikely. I expect that Ethereum will surpass Bitcoin's marketcap eventually, and that will diminish its potential.

querx  ·  886 days ago  ·  link  ·  

Me too. Sideways, ETH is currently being sold (at least officially) via BTC. So a collapse of the BTC system, where we're heading since a longer period of time, would significantly impact Ethereum's use, development and deployment - and that isn't that good.

Maybe we should create a new cryptocurrency - DAO makes it easy [[UPDATE: to lose your money]] ...

Oh, DAO is ETH based, ETH is BTC based, it follows: all DAO-related currencies are BTC dependent in the current stage.

Ethereum should become independent (if it wants to survive) but not take over the role of Bitcoin. It wasn't intended for it in the first place.

mk  ·  886 days ago  ·  link  ·  

    Sideways, ETH is currently being sold (at least officially) via BTC

I think this was true for several months, but increasingly less so. I know a few people that just went fiat to ETH, and the big exchanges support it.

As both ETH and BTC rose over the last several days, I think it's evidence that they are less inversely correlated than before.

joelthelion  ·  888 days ago  ·  link  ·    ·  parent  ·  post: Multi-scale Multi-physics Heart Simulator UT-Heart

This is pretty cool, and looks pretty realistic, too (I did my PhD in the domain). The problem with simulations like these is that they often look cool but fail to accurately reproduce the actual motion of the heart.

Now if this simulator can be tuned to reproduce the heart of individual patients, it would be a great tool for studying the impact of various interventions and deciding on the best course of action.

joelthelion  ·  890 days ago  ·  link  ·    ·  parent  ·  post: Copper is key in burning fat

Of course not ☺️

joelthelion  ·  891 days ago  ·  link  ·    ·  parent  ·  post: Copper is key in burning fat

If you get diabetes from obesity, it's too late to reverse that condition, so it's a bit similar. Same thing if you get a heart attack or a stroke. Anyways, it was just an example. My point is that the fact that there is an existing way to fight obesity doesn't mean we shouldn't try to make that process easier.

vile  ·  890 days ago  ·  link  ·  

Okay, sure, but people also shouldn't give up on currently known to work methods just because there may or may not be an easier way at some indeterminate point in the future.

joelthelion  ·  890 days ago  ·  link  ·  

Of course not ☺️

joelthelion  ·  891 days ago  ·  link  ·    ·  parent  ·  post: Copper is key in burning fat

You could apply the exact same reasoning to lung cancer research. Just because many people could avoid getting lung cancer by not smoking doesn't mean we shouldn't look for a cure.

Note that I am not defending all the fat people who make up excuses for being fat. But I'm all for making it easier to lose weight and stay fit.

vile  ·  891 days ago  ·  link  ·  

You don't reverse lung cancer by quitting smoking, though. If you fix your diet, you'll get to a healthy weight.

joelthelion  ·  891 days ago  ·  link  ·  

If you get diabetes from obesity, it's too late to reverse that condition, so it's a bit similar. Same thing if you get a heart attack or a stroke. Anyways, it was just an example. My point is that the fact that there is an existing way to fight obesity doesn't mean we shouldn't try to make that process easier.

vile  ·  890 days ago  ·  link  ·  

Okay, sure, but people also shouldn't give up on currently known to work methods just because there may or may not be an easier way at some indeterminate point in the future.

joelthelion  ·  890 days ago  ·  link  ·  

Of course not ☺️

joelthelion  ·  891 days ago  ·  link  ·    ·  parent  ·  post: Copper is key in burning fat

I agree with you, but we also know for a fact that it's harder for some people to lose weight than for others. This might be part of it and that's why I think that's why this research is relevant and exciting.

OftenBen  ·  891 days ago  ·  link  ·  

    it's harder for some people to lose weight than for others.

Life isn't fair, and in my opinion research like this takes away from the core problem of 'You have a shit diet' because it gives an out. 'I don't have enough copper in my body to lose weight like you do.' or 'I don't have enough copper, that's why I'm fat' instead of 'I like to deal with stress by eating a block of cream cheese dipped in cane sugar every other day'

WHY someone has a shit diet is a very important question. Much more important than the copper content of their blood/various tissues.

Devac  ·  889 days ago  ·  link  ·  

    WHY someone has a shit diet is a very important question. Much more important than the copper content of their blood/various tissues.

From what I have gathered in my experience, and putting myself as an example, it's because people have no idea that eating healthy and doing own meals takes less time when given some practice and guidelines than eating unhealthy. Before I have started making all of my meals I was one of those "I'll just buy a frozen pizza" types.

Then I have discovered that by the time I'm waiting for oven to heat up I can make myself a god-damned spinach + veggies pot pie for next three days! For almost the same amount of money that I would spend for one frozen pizza. Maybe I'm just blessed by shitty oven in the dorm's kitchen, but it was a shocking discovery that largely made me pursue various cooking techniques and recipes. Right now I can make myself food for a week in advance for, comparatively, pennies. So it saves time, money and turned out to be much more healthy.

Perhaps actual cooking lessons in schools that focus on making fast day-to-day meals could be more beneficial than some of the alternative public campaigns. I never had any weight to lose (if anything I'm a bit underweight), but I do feel better ever since I have stopped eating crap.

joelthelion  ·  891 days ago  ·  link  ·  

You could apply the exact same reasoning to lung cancer research. Just because many people could avoid getting lung cancer by not smoking doesn't mean we shouldn't look for a cure.

Note that I am not defending all the fat people who make up excuses for being fat. But I'm all for making it easier to lose weight and stay fit.

vile  ·  891 days ago  ·  link  ·  

You don't reverse lung cancer by quitting smoking, though. If you fix your diet, you'll get to a healthy weight.

joelthelion  ·  891 days ago  ·  link  ·  

If you get diabetes from obesity, it's too late to reverse that condition, so it's a bit similar. Same thing if you get a heart attack or a stroke. Anyways, it was just an example. My point is that the fact that there is an existing way to fight obesity doesn't mean we shouldn't try to make that process easier.

vile  ·  890 days ago  ·  link  ·  

Okay, sure, but people also shouldn't give up on currently known to work methods just because there may or may not be an easier way at some indeterminate point in the future.

joelthelion  ·  890 days ago  ·  link  ·  

Of course not ☺️

joelthelion  ·  891 days ago  ·  link  ·    ·  parent  ·  post: Microsoft Finds Cancer Clues in Search Queries

Maybe for more cureable diseases it would be justified, though?

jadedog  ·  891 days ago  ·  link  ·  

I'm skeptical. In first world countries, if the disease were more curable, there are already medical test recommendations. For instance, people take routine breast cancer, testicular cancer and colonoscopy screenings. Even there, there has been some findings that the higher rate of testing can create more problems with false positives and the test itself than is really saving lives at the margins. In countries without a national health insurance, money can be an issue to get care as well.

In some parts of the world, even if you knew you had something, there's not much you can do about it. Money is often an issue as well as the availability of medical facilities.

There would have to be some kind of showing that alerting people would be doing more good than harm.