a_square_peg

14 ms·

91.
by a_square_peg 4y ago
Yeah - I work with Python for my main work dealing with weather & climate data but things still throw me off. It feels like every example I run into is an easy to follow "20-steps + 10 packages + 64GB cloud server instance" pr
92.
by a_square_peg 4y ago
Read books - that's really the only consistent way that I know of and which I always recommend to others, especially to young people. Aim to read one book a week - that'd be 500 books in 10 years. Stay away from absorbing informat
93.
by a_square_peg 4y ago
That's fantastic - congrats!
94.
by a_square_peg 4y ago
> Differences within the populace with regard to risk tolerance and perception doesn't tell you anything about systemic issues. How we deal with it as a society does though - some people don't mind going fast but we as a societ
95.
by a_square_peg 4y ago
Maybe slightly off-topic, but I think this is a part of a larger trend of a declining empire - its inability to solve problems that have already become non-issues in other societies. From big issues like healthcare to small stuff like getti
96.
by a_square_peg 4y ago
Yikes - that reference is some dentists-having-opinions-on-brain-surgery conspiracy stuff. I've seen these before and really not sure why people who study rocks have strong opinions on climate science. A few points: - Reanalysis data i
97.
by a_square_peg 4y ago
Here's something similar, created from the same dataset (ERA5): https://climate-explorer.oikolab.com - I created this to show how temperature has changed for any arbitrary location of interest. Working on adding CMIP6 proje
98.
by a_square_peg 4y ago
I run weather data API (oikolab.com) focusing on historical data. Take a look at weatherdownloader.oikolab.com or climate-explorer.oikolab.com for examples and feel free to reach out if you'd like to try playing with the data for visua
99.
by a_square_peg 4y ago
Holy cow, this is amazing. Absolutely fantastic!
100.
by a_square_peg 4y ago
I tend to think most of engineering is rather mundane but potentially very rewarding. I hate meetings and overly complicated project coordination just as much the next person but there is also danger in wanting to do ‘cool stuff’ - imagine
101.
by a_square_peg 4y ago
Perhaps it's just a series of dopamine hits but there is some strange feeling of being satiated from consuming new information while browsing. It's only when I switch off I realize that there is no substance to it or even worse, I
102.
by a_square_peg 4y ago
I think the tools may have become too good. There are so many different ways to build web services and the hardware (CPU/GPU/RAM/network bandwidth) and the software (OS/Nginx/Python/PHP etc.) have become so goo
103.
by a_square_peg 4y ago
This is a fantastic list - thank you for sharing.
104.
by a_square_peg 4y ago
Oh wow, I did not know that. I had only been aware of CANDU reactors and had the impression that there wasn't much else going on in Canada.
105.
by a_square_peg 4y ago
I really don't know enough about this to know if it's a good idea or not (it sounds like it could be quite useful) but in any case I'm very glad to see Canada doing something new in technology.
106.
by a_square_peg 4y ago
Yeah - I meant not seeing the clickbaity titles. Even major news outlets like BBC and NYTimes are becoming like this which is frustrating.
107.
by a_square_peg 4y ago
I wish there was a click-bait headline filter. Too many news headlines are either hypothetical question with no real information content, telling me to re-evaluate everything that I know, something sensationally misleading designed to make
108.
by a_square_peg 4y ago
More and more, I'm acknowledging that finding credible information online that requires any depth is a futile exercise. The return on time spent is horrible and I also realize that more often than not, I don't have the time and kn
109.
by a_square_peg 4y ago
I agree 100% - I think this ability to identify the bottleneck for the system becomes more important as you move up the rank because you don't need to be the one doing it but you do have to know what needs to be done.
110.
by a_square_peg 4y ago
Very true. This is why I think 'Thing Explainer: Complicated Stuff in Simple Words' (Randall Munroe) is so brilliant. I think being able to communicate at the level your audience is an under-appreciated skill.
111.
by a_square_peg 4y ago
I agree it's not a straightforward thing - it's just that in many cases people seem to be hiring a violist but screening them with live sessions of Guitar Hero.
112.
by a_square_peg 4y ago
Would really like to see one day a post about 'when hiring developers, read their CV and have a technical discussion about their past work in relation to the role required' becoming a thing.
113.
by a_square_peg 4y ago
> A prerequisite for generalization is the ability to take something high-dimensional and reduce it to a lower-dimensional representation to allow comparison and grouping of concepts. I've been thinking that it might actually the ot
114.
by a_square_peg 4y ago
Another point about social media - it’s structurally easier to say things (done with a single ‘re-tweet’ click to thousands of people) but increasingly difficult to listen (you still have only one thing you can focus), and even harder to ca
115.
by a_square_peg 4y ago
This is an excellent article - thanks for sharing. This decline of shared common reality and social trust has been both frightening and frustrating for me.
116.
by a_square_peg 4y ago
I feel like there is a lot of condescending tone in the article that's really not helpful to many people who are not as versed in the topic. Maybe like a wine connoisseur who can't give a simple answer to what wine he might recomm
117.
by a_square_peg 5y ago
Yep. Perhaps what I should have written is that much our knowledge is tacit in nature. We implicitly understand that things we can sense are limited projections of the real world and are able to derive mental model of what that reality migh
118.
by a_square_peg 5y ago
I’ve been wondering about the limits of data-centric approach – there seems to be this implicit notion that more data equals better performing ML or AI. I think it would be interesting to imagine a point of diminishing return on additional
119.
by a_square_peg 5y ago
I have a feeling that interview processes have become like dating apps - all these leetcode tests and 7+ interviews, however they're justified, are just filtering systems to deal with the sheer number of candidates. Just like dating ap
120.
by a_square_peg 5y ago
Maybe a slightly less extreme analogy would be highly processed food vs. traditional food... I feel like I'm just learning that white bread isn't really healthy and is nutritionally deficient after many years of vigorous consumpti

More ›