« Highly Recommended | Main | Quote o' the Day: Orhan Pamuk »

Friday, 02 November 2018

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

An interview with Beth Moon. Cheers.
http://blog.photoeye.com/2018/04/interview-beth-moon.html?fbclid=IwAR0rsaJMVz5s77MW8ymCea3oIviQZ_0ywMiW4u3WNLF4SroRx3ZNm2Z9CpI

Many thanks! This is all great stuff I look forward into digging into, after I finish my data analysis report comparing the X-H1 vs the X-T3's AF performance for Fujifilm NA for motor racing photography. I'm doing 2-proportion tests, and calculating Z-scores and p-values, with alpha set to 0.05 (provides a 95% confidence level)

I won't disclose numbers and stats, but the X-H1 has been holding its own against the hot new X-T3 very nicely in a tough, demanding use-case. Better than the YouTube talking heads and pundits would have one believe (but then again, none of their anecdotal "testing" are properly controled experiments with statistical analysis, so they have virtually no know way to know with accuracy).

Using the fantastic new XF200mm Fujinon didn't hurt one bit, either. 'Wot a lens!, as the Brits might say.

I look forward to your thoughts on the X-H1, too.

Have a good weekend, Mike.

Oh, my, thank you for the link to the remembrance of Ara Guler. I had seen the headline, not clicked through, and of course lost it in the fire hose that is the news cycle.

The photos are striking. The flying birds over the barge took my breath away.

That looks like a beautiful book, trees are great subjects for all forms of photography... I was just yesterday looking at the hoops some folks jump through to make "carbon prints", but boy are the results great!

In the corner of the room in which I sit, I have a piece of a very old tree, no tree was hurt in the acquisition, it was just lying there when I went to an old priory garden in Staines, Middlesex around 30 years ago, just to visit the piece of history that it is.

It is called the Ankerwycke Yew (it is pictured on the net) and it is reckoned to be 2500 years old, so... older than Christ.

In my view just as important though and frequently forgotten, by the PTB, it is also said to be the tree under which King John was made to apply his seal to the Magna Carta in 1216. It was already more than 1500 years old.

Despite the fact that the Pope decreed this treaty null and void, the Americans and many here in Britain still regard it as the basis of the compact between people and state, it is the basis of the US Constitution and the basis of our "Common Law" as opposed to the plain wrong Napoleonic Corpus Juris that we are now obliged to cleave to.

The difference is simple.

Under the former, everything is legal, unless legislated against.

Under the latter, nothing is legal, unless legislated for.

It is why I have a little piece in my study, and why I get so excited about people that are dismissive of our decision to leave the EU.

Just as a matter for Americans, who paid homage with a statue at Runnymede, the actual spot is about a quarter of a mile away on the north side of the river, but the honour paid is appreciated. The point is that momentous occasions were almost always marked beneath ancient trees.

The courts were always held beneath these trees, and the upstart Christianity sited their churches where celebrated ancient yews stood, since they were already gathering places. Britain is covered in these trees, but only a few are as old as this one.

Yes sir, trees are important, and this tree probably features in Ms. Moon's book... If it doesn't it should, it is a beautiful thing.

What wonderful photographs! I am a Times subscriber, but had missed the Guler piece. Thanks for highlighting Mike.

Ooh Beth Moon is one of my favourites! I went to England for vacation this year and I tried to map out where some of the ancient trees are so I could see them for myself. I did get to see one of them.

Her books regularly grace my coffee table in a serendipitous rotation based on when I can get them from the public library. When I finally settle on which one I like the best, I'll have to order it so it can be in the house permanently (trusting, of course, that I can actually still buy whichever one I choose!).

"I'm doing 2-proportion tests, and calculating Z-scores and p-values, with alpha set to 0.05..."

WOW!!! I have absolutely no idea whatsoever what any of that meant- but uhhhhh... truly impressive!

PS- NOT, meant as a put down to anyone- other than myself...

@Stan B: A 2-proportion test is a class of statistical tests called hypothesis tests for measuring 2 sets of proportions (aka percentages). It tells you whether the proportion of two data sets are statistically the same, or different.

Hypothesis tests work in a pretty simple way. They work by creating two hypotheses: One hypothesis, the null hypothesis, says the two data sets are the same. In this case, it means the proportions or percentages of in-focus frames from each camera are the same. The second hypothesis statement, called the alternate hypothesis, says there IS a difference between the two, i.e., one camera produces more in-focus frames than the other, and that difference is statistically signficant.

The way hypothesis testing works is the null hypothesis is accepted to to be true (i.e., no difference between cameras) unless the data says otherwise (in which case, you reject the null, and accept the alternate hypothesis). Because of the practical consequences of making a wrong decision can be real or impactful, you want to have some confidence if you reject the null and accept the alternate hypothesis, which depends on how much risk you are willing to accept of making the wrong call. The standard is to accept a 5% risk of making the wrong decision (i.e., saying they are different when the truth is they are not). This risk of making an incorrect decision when the truth is otherwise is called "alpha risk". So, if you want to have a 95% confidence you're making the right call, you have a 5% risk of making the wrong call. That is what I was referring to when I said "alpha was set to 0.05" (i.e, 5%).

In this case, I am counting the number of in-focus frames out of a total number of frames for each camera. What I want to know is: is the percentage of in-focus frames for the X-H1 the same, or different, compared to the X-T3?

Because proportion tests are simply counting test results e.g., good/bad, yes/no, etc., you need fairly large data sets to have confidence in making an inference if the proportions are the same, or different.

The Z-score is the statistical value that is calculated by the 2-proportion test; it produces a number, called the p-value that is the probabillity of rejecting the null hypothesis if it in fact true.

So, if the tests come back and say that the proportion of in-focus frames is different with statistical signficance, I have 95% confidence that those results are, in fact, true.

Just as an aside, the Bodhi tree under which Buddha attained enlightenment is over 2,500 years old. Prince Siddhartha, later known as Gautama the Buddha, is said to have been born in the 6th century BC.
https://www.upi.com/Buddha-tree-alive-and-healthy-at-age-2500/50861342821286/

Pardon my skepticism, but "I'm doing 2-proportion tests, and calculating Z-scores and p-values, with alpha set to 0.05 (provides a 95% confidence level)" boils down to saying that the commenter will compare two sets of numbers while assuming that each set has a Gaussian distribution and hopes to find that the sets are different with 95% probability (that's certainty minus .05). That will make one set fall to the right of the other, but we have hear more about how the sets of numbers reflect our possible satisfaction with either camera in order to know if "to the right" means "better for me."

Beautiful and inspiring work. I love the tonality and grain structure in Guler's work. To me, that is what B&W should look like. It makes digital B&W seem sterile and lifeless. Is that the result of nostalgia or is it simply and demonstratively more beautiful?

Stephen J: great comment. I'll also add the Tree of Life and the Tree of Knowledge.

Stan B: you speak for me too!

The older I become, the less I know or understand - or care.

:-)

Rob

Stephen J: if your later life had been invested in living in the EU, your children additionally schooled in languages with a future in highly-paid work in Europe a very distinct possibility, then perhaps you wouldn't get so excited over lemming symptoms.

As ever, it's the sons (and daughters, in the interests of fairness) carry the penalty for the sins of the fathers.

:-)

Rob

There's a documentary about Ara Gruler available for steaming on Amazon. It's free for prime members.

The poet of Istanbul.
He made photos that sing.
Thanks for pointing this out.

The Dianne Arbus of trees.

@Stephen Scharf

I can't help but believe anyone who can describe that in such detail and accuracy, even if you were lying to my face- not that I in any way believe you are. You sound like one serious dude, who seriously knows what he speaks of! I was just emphasizing my own ignorance...

@Bob Campbell

It's about the only upside to it...

Not meaning to weigh in on Brexit issues, but the post comparing the common law to the Napoleonic code has it backwards. The common law of England and the United States historically recognized "common-law crimes" defined in judicial decisions, crimes which were not the product of legislation. Under civil law systems, such as the Napoleonic Code,judge-made crimes could not exist, and only statutes could make acts criminal. Most if not all States of the United States, and United States federal law, have since abolished common-law crimes, and make conduct criminal only through legislation.

Re: Beth Moon’s work, it’s lovely. Work initiated by, and concentrated on, an idea, rather than a sight, is nearly always the more durable and rewarding. This work is certainly that. I personally usually prefer to initially see such images online and then later as art (PP, in this case) prints. Such carefully made olde chemical prints can be very seductive objects in themselves and often mask an otherwise dull, untalented eye. (Although it’s most often the medium that gallerists find most salable, withe subject often only a potential detraction. )

Relatedly, also in 2014 I purchased a photo/essay book titled The Oldest Living Things in the World by Rachel Sussman. It’s a broader work contemporary to Moon’s that’s also available on Kindle. I certainly appreciate the effort that was invested in such a project. But this was more of an illustrated eco-list documentary whereas Moon’s project is eco-art.

Two tidbits:
The first test for comparing means is called Students's T test. It was developed to compare batches of Guiness Beer. Student was a psuedonym because Guiness did not want him to associate the T-Test with Guiness.

The other is too bad, Beth did not start this project in 1964. Then she could have taken a picture of the *real* oldest tree:

https://www.smithsonianmag.com/smart-news/how-one-man-accidentally-killed-the-oldest-tree-ever-125764872/

The comments to this entry are closed.

Portals




Stats


Blog powered by Typepad
Member since 06/2007