How Pure Is This Cup of Joe? | Hacker News Reader

How Pure Is This Cup of Joe?

hackaday.com

140 pointsszczys21 comments

> Finally, I used LibreOffice to tabulate and graph the data, and perform t-tests. T-tests are simple statistical tests used to determine the probability that a null hypothesis (H0) is correct. If it is less than 5% probable that the null hypothesis is correct given our data, we will reject the null hypothesis and consider our alternate hypothesis (HA) to be correct.

Obligatory stats nerd: that's not what a p-value means, that reverses it.

Reverse isn't the right word here, either. Better to restate the definition: the p-value is the probability of observing the data (as represented by its aggregate statistics), under the condition of the null hypothesis.

I say 'reverse' because that's what it is formally: confusing P(data|null) with P(null|data). It reverses the conditional probability.

Reversing the arguments of the conditional probability, yes, but just saying "it's the reverse" is a bit ambiguous.

> If it is less than 5% probable that the null hypothesis is correct given our data, we will reject the null hypothesis and consider our alternate hypothesis (HA) to be correct.

You could be reversing the 5% to 95%, could be reversing null hypothesis to alternate hypothesis, could be reversing reject to accept, etc. Too many moving parts. P-values are so often misinterpreted, it's best to err on the side of specificity.

My biased opinion is that the only reason why p-values are popular is that it is a concrete question whose answer can be confused with the impossible question that we really want an answer to.

If you object to the confusion, you've just taken away the main reason why it is popular.

Does anyone know what the fish oil additive is used for?

I was recently researching how Trader Joe's coffee is so inexpensive and found some reviews noting a "fish sauce" flavor and a "fishy" flavor in two TJs coffee products.

https://www.thrillist.com/culture/best-trader-joe-s-coffee-t...

[edit] Here's a photo of the back of a bag of his "reference" Trung Nguyen S blend. It explicitly says that it has flavorings in it, including "coffee flavor". Omfg.

https://imgur.com/a/KeSYq

---

I guess the guy had fun, but the conclusion is a few giant leaps away from the actual results. Many of his decisions and assumptions were entirely arbitrary, not well substantiated, and lacking just about anything approaching rigor.

His reference coffee is "A blend of Arabica, Robusta, Chari (Excelsa) and Catimor beans"[1], roasted in a special style, and brewed in one particular way with no mention of temperatures involved or any other variable that alters the outcome of brewing coffee.

Trung Nguyen coffees are as far as I know usually adulterated with chocolate and other flavoring either during or after roasting. This is not a bad thing. It smells and tastes goddamn fantastic and they are my favorite coffees in the world. But it's not exactly a good reference.

His reference chart includes only two of those bean types and specifically references green beans, not roasted beans, not butter roasted beans, and not beans roasted at any particular range of temperatures or durations.

He doesn't indicate that a reference coffee was chosen based on lowest wholesale cost, which might have been a reasonable decision factor for a street vendor, but rather just that it seems to be commonly available on the consumer market without regard for composition or lack of monopoly.

His reference not-coffee is only one of plausibly dozens of different not-coffee substances.

His reference not-coffee solution is based on approximately tasting like terrible coffee. Whaaaaaat.

And even if you choose to ignore any or all of the above, nothing about his method tests for coffee. Coffee is generically an extremely complex blend of varying volatile compounds, but specifically it is an infinity of different complex blends of varying volatile compounds. And his process doesn't actually test for any of them. It tests for some secondary characteristic that one mixture possesses, ability to change the color of iodine, but that one arbitrarily different other chemical does not possess to the same degree under the assumption that all real coffee changes the color of iodine the same and that all non-coffee also changes the color of iodine the same. These are exceptionally ridiculous assumptions.

[1] http://www.trung-nguyen-online.com/trung-nguyen-coffee-order...

> Overall it sounds like a reasonable test for the amount of coffee in something

No, it really does not.

This is what happens when people think that "science" is synonymous with measuring shit.

Why stomp all over somebody's fun? I enjoyed reading it.

Because awareness of the gap between measurement and conclusion might just be the most important thing in the world, and people and governments fail spectacularly at it all the time with significant lasting effects on society.

> the gap between measurement and conclusion might just be the most important thing in the world

What happened to love, compassion and humility?

Trying to get to the factually right conclusion is an expression of humility, because it allows admitting that our initial expectations were wrong.

Using the right conclusion to properly guide our actions is necessary for compassion and love -- if you truly with to benefit others, then your actions must be correct and guided by the right answers.

The challenge is updating other people's beliefs without being seen as an attack.

The price of Bitcoin ?

For anyone reading this who doubts what BugsJustFindMe said here, as an example I invite you to examine how the government measures the loudness of broadcast commercials compared to the programs that surround them. We had seventy years of blasting people with commercials, all because the way of measuring the loudness does not match the actual "loudness".

Take that anti-cynical perspective back to Reddit thank you, here in Hacker News we have no time for such leniency.

===

On a serious note, I enjoyed reading it. I think it's clear(ish?) that the approach taken by the author leaves a little to be desired, he seems to be modest and quite aware of this.

It's no different to Myth Busters really. They use the scientific process and rarely express them selves past a sample size of one. But for the purpose of advocating science and it's methods, I think it's good. If it's not well... I'll leave you all to stick to reading Journals?

Because that somebody is pretending they have useful quantified results, and they don't.

And "man plays chemist with iodine for a few hours" probably wouldnt make the front page.

Should be using the ISO Standard Coffee :)

ISO 24114:2011 https://www.iso.org/standard/44704.html

Catimor is a variety of arabica, not a different species like robusta. Excelsa seems to refer to the Colombian coffee grade which may include different arabica varieties.

I thought the rumour here says that the fake coffee is made of roasted soybean (to add bulk), quinine (bitter), gelatin (thickness), not ready made artificial flavourings as was tested in the article. Although I highly doubt that there is no coffee content in street vendor shop, as it doesn't make sense financially (Vietnam is the 2nd largest coffee producer), so the rumour is the same as "Chinese plastic rice" or "Chinese fake eggs" IMO.

Need to read the article, but after recent occasional experiences with an associate's, I've been wondering whether the flavor profiles I experience with K-cup coffee servings are due to flavoring agents. The flavors have that kind of uniformity and intensity. And I am rather suspicious of their coming from a package of pre-ground coffee with a shelf life of months or years and manufactured/sold at the price point their are finding.

You should test the major chain coffees

I wonder what McDonald's uses to flavor their coffee. In Canada it has a very distinct taste unlike any coffee I've had. It can not be pure coffee in my opinion.

It claims it is 100% Arabica beans but it is so consistent and distinct. I wonder I fit does the 100% pure orange juice "trick" by processing parts of the Arabica beans into flavor chemicals and reading it.

Reply on news.ycombinator.com