Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
martingoodson
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
91.
▲
For elite footballers, the effects of Covid-19 linger for months
(economist.com)
25 points
by
martingoodson
5y ago
|
1 comments
92.
▲
by
martingoodson
5y ago
I’ve made this comment before: If anyone wants to start eating wild mushrooms and doesn't want to poison themselves I recommend this approach: 1. Choose one edible mushroom species that is often found in your region and is difficult to
93.
▲
by
martingoodson
5y ago
One of the great writers on science. This is from ‘The Diversity of Life’: “In the Amazon Basin the greatest violence sometimes begins as a flicker of light beyond the horizon. There in the perfect bowl of the night sky, untouched by light
94.
▲
by
martingoodson
5y ago
I met a coconut picker in Cuba who had survived a coconut falling onto his head. It landed rounded-part down, rather than pointed-part down: doctors thought that was probably why he survived. He told me his vision temporarily went completel
95.
▲
by
martingoodson
5y ago
Just wanted to say I was taught to program by a volunteer teacher in a youth club. I was ten years old. It changed the entire course of my life. This work makes a difference!
96.
▲
by
martingoodson
5y ago
The consumer price index does include food and energy. It's in the main table of the post we are discussing. It's not clear why you are not 'buying' them. Food...................... .4 .8 .7 .4 .9 .
97.
▲
by
martingoodson
5y ago
This ‘bad submission’, despite many upvotes and significant engagement, is flagged and removed. In favour of these submissions on the front page: ‘THOSE BULLET EFFECTS IN TERMINATOR 2 WEREN’T CGI’ ‘Your Red Snapper Filet Might Have Been Cau
98.
▲
by
martingoodson
5y ago
Clearly a number of others do feel that. I wonder whether those people would also criticise the Tank Man image [1] as not ‘substantive’. Reportage does not need to contain analysis to be effective and valid. [1] https://en.m.wiki
99.
▲
by
martingoodson
5y ago
It is ‘noteworthy’ that the President of the COP26 considers it to have been such a profound failure that he is literally crying on stage.
100.
▲
by
martingoodson
5y ago
Who on Earth told you that climate change mitigation was “off topic for HN”? I posted it because it conveys how deeply f*cked the planet (and the human species) is much better than pages of analysis.
101.
▲
Holding back tears, the COP26 president says he is “deeply sorry”
(twitter.com)
15 points
by
martingoodson
5y ago
|
21 comments
102.
▲
by
martingoodson
5y ago
This experiment was probably falsified. https://statmodeling.stat.columbia.edu/2019/08/20/did-that-b...
103.
▲
by
martingoodson
5y ago
Delta
104.
▲
The UK AI Strategy: are we listening to the experts?
(datasciencesection.org)
1 points
by
martingoodson
5y ago
|
0 comments
105.
▲
The UK AI Strategy: are we listening to the experts?
(linkedin.com)
2 points
by
martingoodson
5y ago
|
0 comments
106.
▲
by
martingoodson
5y ago
‘Where’ is it? Where is homeostasis? Where is the immune system? As tsimionescu suggested, a spatial metaphor is not the only way to understand phenomena.
107.
▲
by
martingoodson
5y ago
The article we are both commenting on is in a scientific journal and is concerned with scientific research into consciousness. I think it’s appropriate to talk about scientific results here.
108.
▲
by
martingoodson
5y ago
The history of Western philosophy has amply demonstrated that investigating the grammar of language is not a useful way to find out how the world works.
109.
▲
by
martingoodson
5y ago
Great. Could you tell us what scientific fruits have resulted?
110.
▲
by
martingoodson
5y ago
Well I typed it, obviously. But my existence doesn’t say anything about whether a ‘hard problem’ of consciousness is a useful way to think about the human mind.
111.
▲
by
martingoodson
5y ago
I agree. This quasi-mystical framing seems unlikely to bear scientific fruits. Is consciousnesses a clearly defined term?
112.
▲
Thinking fast, slow, and not at all: System 3 jumps the shark
(statmodeling.stat.columbia.edu)
64 points
by
martingoodson
5y ago
|
13 comments
113.
▲
by
martingoodson
5y ago
Amoeba are not animals. Animals (metazoa) are multicellular organisms.
114.
▲
by
martingoodson
5y ago
This is called ‘induced demand’. Surprisingly, the opposite happens when you remove road capacity - which is called ‘traffic evaporation’. It’s why taking road space from cars and reallocating it to bicycles is often more successful than p
115.
▲
by
martingoodson
5y ago
The parent comment says ‘while being paid the exact same as someone on the SWE or PM track.’ Not ‘less than a SWE’, as you imply. Why should a data scientist be paid more than a SWE? Because they have to learn several different topics? That
116.
▲
by
martingoodson
5y ago
Learning all this is not really that difficult. No more difficult than a biochemist training in subjects as diverse as organic synthesis (making stuff in test tubes), Raman spectroscopy (prediction of chemical structures using vibrational s
117.
▲
Why is it so difficult to define artificial intelligence?
(evolution.ai)
2 points
by
martingoodson
5y ago
|
0 comments
118.
▲
Why is it so difficult to define artificial intelligence?
(evolution.ai)
2 points
by
martingoodson
5y ago
|
0 comments
119.
▲
by
martingoodson
5y ago
I'm sure it's a coincidence that Fujitsu was also heavily involved in the NHS IT fiasco which cost the NHS £10B. 'the biggest IT failure ever seen'. The Fujitsu UK chairman is also a large Conservative party donor of cou
120.
▲
by
martingoodson
6y ago
Can somebody explain the advantage of Streamlit over sharing a Jupiter notebook with interactive widgets? Are data scientists sharing models and data a relevant use case?
More ›