Interesting Links for 09-05-2021

[andrewducker]

There's no such thing as a tree (phylogenetically)

(tags:trees ontology nature evolution )

The mandate for a Scottish independence referendum

(tags:Scotland independence )

Interesting changes in the Blue Wall English Remainer areas

(tags:UK politics Europe )

People aged 105 and over have more efficient DNA repair

(tags:age DNA genetics )

"I wrote the book on user-friendly design. What I see today horrifies me"

(tags:design age fail )

Comments

[rhythmaning]

It's not that there's no such thing as a tree. It is rather, as I have tried to explain to Frankie several times, there are a very many different ways of being a tree.

(one of my favourite books from my botany degrees was Halle, Oldman and Tomlinson, "Tropical Trees and Forests - An Architectural Analysis" which details about three hundred, I think and much to my surprise doesn't appear anywhere on the internet other than in the Springer catalogue.)

Also, I've forgotten the LJ code for tagging someone, because I think this might amuse Frankie...

[andrewducker]

Hence the word in brackets!

[rhythmaning]

Indeed!

As an aside, one of my favourite trees... Is the banana! Which, like bamboo, is really a grass... (Banana taxonomists would tell me off for that. But they're nearly grasses...) Banana trees are very easy to grow: they reproduce by adventitious buds, so they're easy to clone. As a student, one of the other PhDs used bananas in some of his physiology experiments, after which he'd get rid of the plants. So we all had banana plants on the windowsills. Not trees, though - maybe saplings...

[danieldwilliam]

The tree article reminds me of three things.

1) a lot of our problems as humans is that we looked at the world and grouped and organised things it by observation before we understood the underlying mechanisms at play and we've been playing catch up (badly) with how things actually are for a long time.

2) evolutionary processes do not care if they are tidy.

3) I miss Richard Dawkins not being a massive bellend.

[andrewducker]

Agreed on all three counts.

[rhythmaning]

In answer to (1), pre-Linnean taxonomists used classifications which were useful at the time, and most still do. Gardeners want to know what colour a flower will be, regardless of the family. (I simplify, obviously.) Foragers need to know which plants are safe and which are poisonous. "Tree" is a useful thing to identify.

(2) biological processes don't care at all.

(3) Dawkins was always a bellend, even when I was a student. But his work on evolution, particularly "The Selfish Gene", was always interesting. (Although following (2), I feel the need to point out that genes are neither selfish nor good.)

[mountainkiss]

(3) E vero. I was at the college where he taught.

[rhythmaning]

Yes, I thought that was the case.

His ex-wife taught friends of mine at Somerville when they splitting up, and apparently he was an utter bastard.

[mountainkiss]

That's also my understanding.

[mountainkiss]

You had some nice ferns until Talisker started sitting in the pots.

[rhythmaning]

I still have some nice ferns! Which I have had since they colonised pots in the Botanics greenhouse I used. I also had some pots of bracken until I moved back to London...

This means that the ferns I have got - all clones of an original one - are twenty five eyars old...

[mountainkiss]

You should have a party for them.

[jducoeur]

It's not that there's no such thing as a tree. It is rather, as I have tried to explain to Frankie several times, there are a very many different ways of being a tree.

More generally, the words/concepts humans routinely use frequently have little to do with precise categorization. Our brains are very effective, but rather general, pattern-matching engines, so we tend to categorize things in complex, multi-dimensional "this feels right" sorts of ways.

Back in college (so this is back in the 80s), the single most memorable AI class session I took was one where we started class, and the teacher asked a simple question: "Define 'table' for me". We spent an entire hour trying to come up with a logical definition of a table, but every time we would come up with a rule he would immediately demonstrate a counter-example that violated our definitions.

That made a deep impression on me, that trying to impose logical consistency on basic human terms is typically going to be a lose. (And that rule-based AI systems were going to be mostly a waste of time, but it took another 15-20 years for that clue to percolate around the industry.)

[conuly]

People aged 105 and over have more efficient DNA repair

Makes sense. You can reach age 80 through healthy living, but the only way to make it to 100 is to have good genes.

[snippy]

One of my great grandmothers lived to 105, so there's a chance I have these genes.

[andrewducker]

Fingers crossed!

[momentsmusicaux]

I feel the tree diagram needs a comment from the infamous internet person who ranted about children being taught trendy nonsense such as whales being mammals, saying 'IF IT SWIMS IN THE OCEAN, IT'S A FISH!'.