Re:* cures (or causes) cancer
by threedaymonk on Wednesday July 07, 2010 @05:21PM
(#1495518)
I made this - thanks for voting it to the top of HN!
In case anyone's still unclear, it is a tongue-in-cheek dig at the Daily Mail's reporting of cancer stories. It's most definitely not a medical resource. I could probably have done the same with almost any newspaper, but the Daily Mail is uniquely deserving of opprobrium.
I wrote the whole thing one Sunday night and tweeted about it at midnight; by midday the next day, I had successfully crowdsourced the analysis of all the articles. If I'd known how successful it would be, I'd have written a better, more detailed crowdsourcing process.
Re:* cures (or causes) cancer
by joe_the_user on Thursday July 08, 2010 @12:10PM
(#1497789)
I'm truly unamused by your lame thingy.
Sure, you can take a poke at this or that not entirely proven cancer association. But your "tongue-in-cheek" site definitely facilitates the devaluing of the many and subtle cancer associations that have been demonstrated - it is not a constructive contribution to the discussion on this subject. For example, while the case automobile pollution causing cancer might not be air, do you have proof that it doesn't cause cancer? Having "incorrect" as the alternative for each assertion on the Daily Mail is seriously problematic.
Re:* cures (or causes) cancer
by tcdent on Wednesday July 07, 2010 @04:52PM
(#1495409)
It's amazing how many food sources listed in their raw form aren't themselves guilty, but the industrial manufacturing processes and additives used to enhance them are.
Re:* cures (or causes) cancer
by Tichy on Wednesday July 07, 2010 @06:57PM
(#1495817)
What really pisses me off in general is adding useless stuff to things that make said things worse. For example coloring agents in food, but the same goes for superfluous features in software.
Re:* cures (or causes) cancer
by mcantor on Wednesday July 07, 2010 @05:19PM
(#1495516)
Did they add the site's tagline after most of these HN comments were made? It clearly reads, "Help to make sense of the Daily Mail's ongoing effort to classify every inanimate object into those that cause cancer and those that prevent it," which makes it pretty obvious that they're poking fun. Everyone here seems to be reacting as if this site is claiming to be a reputable source of information about cancer-causing materials.
Re:* cures (or causes) cancer
by tcdent on Wednesday July 07, 2010 @05:28PM
(#1495548)
Sure, The Daily Mail, just like any other mainstream news source creates sensationalist headlines to shock you into reading. That doesn't make the studies cited any less accurate. Nor should it prevent us from having an on-topic conversation about something more in-depth.
Re:* cures (or causes) cancer
by nevinera on Thursday July 08, 2010 @06:48PM
(#1499041)
No, applying statistical methods to this instance is unreasonable. Observing that selection bias skews the probabilities involved is perfectly reasonable, and indeed almost necessary to rational discussion.
Re:* cures (or causes) cancer
by pavel_lishin on Wednesday July 07, 2010 @05:47PM
(#1495610)
"We've cured cancer so many times in rodents, if I were a mouse, I'd take up smoking unfiltered Marlboros while working at a nuclear waste disposal site in the Ukraine."
Re:* cures (or causes) cancer
by simon_ on Wednesday July 07, 2010 @05:10PM
(#1495490)
A lot of comments here missing the point... The Daily Mail is a tabloid that is exploiting people's (evident even here on HN) love of hearing about cancer cures/causes regardless of evidence.
Re:* cures (or causes) cancer
by pbhjpbhj on Thursday July 08, 2010 @07:51AM
(#1497064)
Looking at one of tcdent's links (bacon!) it seems pretty well balanced and plainly reported to me:
"However, experts urged caution over using the results of the trials on mice to predict the effect of phosphates on humans.
Professor Stephen Spiro, of the British Lung Foundation, said: 'The authors claim that in mice with lung cancer a diet high in phosphates increases lung cancer growth rates.
'While this may be a relevant observation, it has never been assessed in man. Further study would be required to ascertain any link in humans.'
Dr Kat Arney of Cancer Research UK added: 'Smoking is by far the main cause of lung cancer and quitting is the best way to reduce the risk of this disease.'
They stated the limitations of directly applying the result to humans, they interviewed 2 professionals in the field. Stirling. If you're right this must be an aberration?!
Re:* cures (or causes) cancer
by bruceboughton on Thursday July 08, 2010 @07:59AM
(#1497075)
Most Daily Mail readers are incapable of reading past the headline, so stuffing balance near the end of the article usually elicits the desired effect.
Re:* cures (or causes) cancer
by nkassis on Thursday July 08, 2010 @04:08AM
(#1496799)
This site is quite hilarious. Now can we get a iPhone app that lets me take a picture of an object and tell me my probability of getting cancer according to the daily mail index by using this object?
Re:* cures (or causes) cancer
by adammichaelc on Wednesday July 07, 2010 @04:55PM
(#1495425)
Vitamin D is also really important for immune health. Especially for knowledge workers, who don't normally get enough sun. Recent studies have found that Vitamin D actually activates your T-cells, so without adequate levels, your immune system will act like a depressed dog during a home-invasion and just lay down.
Re:* cures (or causes) cancer
by sajithw on Thursday July 08, 2010 @01:29AM
(#1496578)
Scraping real scientific data from journal articles/headlines published in PubMed would be a lot more interesting and noteworthy.
Not to mention "causing cancer" is a bit of a misnomer. Cancer typically arises when genetic mutations cause a cell to exhibit uncontrolled and sustained division. Our cells experience thousands of mutations a day, most of which are fixed. Some of these inanimate objects may upon being metabolized or coming into contact with a cell, cause mutations. With more mutations, you are at a higher risk for one to go unrepaired and lead to something other than cell death. Colloquially, I understand we refer to certain things as causing cancer but I think the idea of grouping everything into discrete boxes isn't really something worth pursuing in this case.
Re:* cures (or causes) cancer
by jamesbkel on Wednesday July 07, 2010 @04:56PM
(#1495430)
If you read the about page http://kill-or-cure.heroku.com/about, you will see that the source is the Daily Mail and the explanation seems rather tongue-in-cheek.
Re:* cures (or causes) cancer
by te_platt on Wednesday July 07, 2010 @04:55PM
(#1495423)
I didn't even realize it was an attempt at a serious story. "Age causes cancer" sounded like a spoof. Not to mention saying alcohol causes AND cures cancer.
Re:* cures (or causes) cancer
by Terry_B on Wednesday July 07, 2010 @07:37PM
(#1495920)
I have long wanted to see someone attempt to live their life following every piece of advice on how to increase your longevity that appears in trashy current affairs tv shows, bad newspapers etc. Would make a great blog.
Re:* cures (or causes) cancer
by acangiano on Wednesday July 07, 2010 @04:40PM
(#1495362)
I believe that's the point of this site. The tagline is: "Help to make sense of the Daily Mail’s ongoing effort to classify every inanimate object into those that cause cancer and those that prevent it.". The media alarmism and sensationalism are damaging to humanity's fight against cancer.
Re:* cures (or causes) cancer
by coryl on Wednesday July 07, 2010 @04:52PM
(#1495411)
Partially true information can be just as misleading as completely false information.
For example, I looked up "Cannabis"; only 1 entry saying it causes cancer. But I know from other articles on other sites that there are properties that fight cancer. (I don't know if it "prevents" cancer though).
Re:* cures (or causes) cancer
by jodrellblank on Wednesday July 07, 2010 @08:36PM
(#1496053)
Why cite "other studies on other sources" and claim it "needs more data" if you know the point is present data from one source? Not to present an objective look at cancer studies, but to present a filtered look at Daily Mail articles on cancer studies?
In case anyone's still unclear, it is a tongue-in-cheek dig at the Daily Mail's reporting of cancer stories. It's most definitely not a medical resource. I could probably have done the same with almost any newspaper, but the Daily Mail is uniquely deserving of opprobrium.
You can get the code and a database export on GitHub if you're interested: http://github.com/threedaymonk/kill-or-cure
I wrote the whole thing one Sunday night and tweeted about it at midnight; by midday the next day, I had successfully crowdsourced the analysis of all the articles. If I'd known how successful it would be, I'd have written a better, more detailed crowdsourcing process.
Sure, you can take a poke at this or that not entirely proven cancer association. But your "tongue-in-cheek" site definitely facilitates the devaluing of the many and subtle cancer associations that have been demonstrated - it is not a constructive contribution to the discussion on this subject. For example, while the case automobile pollution causing cancer might not be air, do you have proof that it doesn't cause cancer? Having "incorrect" as the alternative for each assertion on the Daily Mail is seriously problematic.
Edit: I can't help myself. Some examples:
Bacon causes cancer: "High doses of inorganic phosphate salts – which are used to enhance the texture and flavour of processed meats..." http://www.dailymail.co.uk/health/article-1102368/Additives-...
Eggs cause cancer: "The chemicals, residue from drugs given to chickens..." http://www.dailymail.co.uk/news/article-50995/Cancer-chemica...
Rice causes cancer: "Cotton production relied heavily on arsenic pesticides. It left residues in the soils which are still there and is being picked up by the rice." http://www.dailymail.co.uk/health/article-444222/Rice-tainte...
Spices cause cancer: "...foreign producers are colouring spices to boost prices and profits." http://www.dailymail.co.uk/health/article-347287/Cancer-chec...
EDIT: Oops... looks like acangiano beat me to it: http://news.ycombinator.com/item?id=1495362
We even both used the term "tagline," which is kind of interesting, since it could just easily have been called a "subtitle" or something.
We were clearly meant to be best friends, Mr. Cantor. :-P
1 - http://xkcd.com/165/
http://www.express.co.uk/ourpaper/view/2010-07-08
No, but it does make it statistically more likely that they are inaccurate. It's known as 'study selection bias'.
Applying statistical likelihood in this instance is stretching the bounds of statistics considerably more than The Daily Mail stretched them.
JunkDNA, http://news.ycombinator.com/item?id=1020444
http://www.youtube.com/watch?v=5eBT6OSr1TI
(Scaremongering cancer headlines near the end here - http://www.youtube.com/watch?v=5eBT6OSr1TI#t=1m51s )
In the same way as the Daily Mail, the BS headlines is problematic regardless of the valid point available when you read the article.
"However, experts urged caution over using the results of the trials on mice to predict the effect of phosphates on humans.
Professor Stephen Spiro, of the British Lung Foundation, said: 'The authors claim that in mice with lung cancer a diet high in phosphates increases lung cancer growth rates.
'While this may be a relevant observation, it has never been assessed in man. Further study would be required to ascertain any link in humans.'
Dr Kat Arney of Cancer Research UK added: 'Smoking is by far the main cause of lung cancer and quitting is the best way to reduce the risk of this disease.'
Read more: http://www.dailymail.co.uk/health/article-1102368/Additives-... [link added by copy-paste]"
They stated the limitations of directly applying the result to humans, they interviewed 2 professionals in the field. Stirling. If you're right this must be an aberration?!
We have an acronym for this sort of thing: FUD.
eg "Immigrants linked to terrorism cancer causing pedophiles"
edit: The above would have to assume the data from the site is reliable, which we can't assume based on the source being Daily Mail.
http://wholehealthsource.blogspot.com/2008/10/vitamin-d-its-... http://www.scientificamerican.com/blog/post.cfm?id=another-r...
Not to mention "causing cancer" is a bit of a misnomer. Cancer typically arises when genetic mutations cause a cell to exhibit uncontrolled and sustained division. Our cells experience thousands of mutations a day, most of which are fixed. Some of these inanimate objects may upon being metabolized or coming into contact with a cell, cause mutations. With more mutations, you are at a higher risk for one to go unrepaired and lead to something other than cell death. Colloquially, I understand we refer to certain things as causing cancer but I think the idea of grouping everything into discrete boxes isn't really something worth pursuing in this case.
Now I have a legitimate reason for making way too much spaghetti bolognese, well beyond it being my favorite meal of course.
Then again, I guess the link's been well established already.
For example, I looked up "Cannabis"; only 1 entry saying it causes cancer. But I know from other articles on other sites that there are properties that fight cancer. (I don't know if it "prevents" cancer though).
Needs more data?
But if this were to have practical application, and the potential to solve problems via the crowd, why not improve it and give it better data?