HNHacker News
TopNewBestAskShowJobs

ovis

615 karma · joined February 21, 2013

submissionscomments
ovis··on Airbnb Adopts Rules in Effort to Fight Discrimination by Its Hosts
What if there was a reputation threshold for a traveller to make an instant booking? Of course it's still not immune to racism or sexism, but it seems that it would provide a decent buffer against it.
ovis··on Karta 0.7 - Geospatial Python
Karta is a package I've been working on for a while now. Originally, it was intended to be a light-weight pure Python module for geospatial data, but it's gradually grown C and Cython extensions in the name of performance.

It doesn't wrap a large C library like Shapely or rasterio, but instead tries to be more general and provide the set of geographical objects and spatial algorithms that I would otherwise need to frequently reimplement.

Anyway, I'd be interested in getting some feedback on it. I use it pretty extensively in my own work, and would like to see it be useful for others.

ovis··on Mongolia is changing all its addresses to three-word phrases
It changes every three meters, so perhaps it depends on which room in the embassy?
ovis··on Mongolia is changing all its addresses to three-word phrases
It does if you're a visitor trying to find an address in the local language.
ovis··on Should Mapcodes Replace GPS?
Technology is great, but there should be graceful fallbacks.
ovis··on WhatsApp, Used by 100M Brazilians, Shut Down Nationwide Today by a Single Judge
Do we know for sure what cases this particular order is connected to? Because WhatsApp has been encrypting Android communications since 2014 [0].

[0] https://www.wired.com/2014/11/whatsapp-encrypted-messaging/

ovis··on The Geographical Oddity of Null Island
But you load a map with incorrect data. Wouldn't a good JS error be much more explicit? I do a lot of geospatial data crunching, and the thought of having wrong but valid data injected into my pipelines makes me uneasy.
ovis··on The Geographical Oddity of Null Island
Sure, but why not raise an exception? Or return a separate flag? Or even do something like (999, 999) (clearly not valid, at least for latitude)?
ovis··on The Geographical Oddity of Null Island
What geocoders use (0,0) as an error condition?

Outputting a valid value as an error flag seems like a pretty bad idea - is this actually common?

ovis··on How Many Decimals of Pi Do We Really Need?
Yes, perturbation methods are still taught and are still recognized as important. The first math course I took during my graduate degree (Intro analytic methods) covered it, for example.
ovis··on Millennials Flee Vancouver for More Affordable Cities
The GP claims most people on HN, which seems likely.
ovis··on Why Digital Maps Are Inaccurate in China
The article implies that the source of trouble is the GCJ-02 datum.

Is this really such a major problem though? In North America, both the NAD27 and NAD83 datums are still common. NAD83 nearly matches WGS84, but NAD27 can be quite off. Datum transformation is a well understood topic, so what's so special about China?

ovis··on America Is Flint
The politician who raises taxes or shuts down water delivery may be replaced at the next election by the one that doesn't and keeps the lead pipes.
ovis··on From Python to Lua: Why We Switched
The article did mention a desire to use netifaces, but I didn't quite understand why the author thought they couldn't.
ovis··on Jupyter Notebook User Experience Survey
For me, it's nice for interactive code and visualization development. Particularly when parts of a script are slow running, being able to do it once and then iterate on a dependent part in a notebook cell saves a lot of time.

Once I'm done, I can document it inline with markdown and latex.

ovis··on Why I Still Use Python for High Performance Scientific Computing
That seems very unfair. I have no reason to think the author doesn't understand that Python libraries are built on carefully written C and Fortran, and based on the article they seem to be a perfectly competent programmer.
ovis··on Why I Still Use Python for High Performance Scientific Computing
But in those cases, the Python programmer could just wrap the C library in Python, which I suspect is often a lot easier than starting from scratch.
ovis··on NNSA, Nvidia to create an open-source Fortran compiler front-end for LLVM
Yeah, when I need neighbouring pixels I extract overlapping chunks. As far as the edge of the scene - there aren't any perfect solutions that I know of. You could try to fudge it by modifying the algorithm to handle the edges with fewer pixels, or perhaps use data from different scenes as an approximation. Or you may simply lose the data there.
ovis··on NNSA, Nvidia to create an open-source Fortran compiler front-end for LLVM
Need more details on what you're doing, but many simple operations used in satellite image analysis are embarrassingly parallel. You could, for example, read in chunks or strips using GDAL, distribute them across multiple threads or cores, and get some speed-up. If much of the time is spent in numpy rather than Python, I think it should be non-blocking so the GIL won't get in your way.
ovis··on Canada's Muzzled Scientists Can Speak Freely Again
Not usually illegal, but grounds for disciplinary action or firing. There are many cases of the government punishing employees for inappropriate speech while on the job. Garcetti v. Ceballos (2006) is a Supreme Court decision finding that some speech of public employees is not protected.

And then there's the Hatch Act (1939), which prevents most federal employees from engaging in activity that could be seen as political.

ovis··on Canada's Muzzled Scientists Can Speak Freely Again
Even in America there are restrictions about what federal employees may say to the media. Freedom of speech is different when you are representing an employer, although the restrictions on Canadian scientists were egregious.
ovis··on According to NASA, Antarctica is actually gaining ice
Previous discussion of the NASA release: https://news.ycombinator.com/item?id=10487680
ovis··on According to NASA, Antarctica is actually gaining ice
Even prior to this, it was thought that Antarctica was only a small fraction sea level rise at present. Thermal expansion of sea water, melting mountain glaciers, and Greenland appear to contribute far more. The latest IPCC report (Chapter 13, table 13.1) has some good numbers representing the best we knew a couple of years ago.

The impact of this study is that the small mass loss attributed to Antarctica may be a small mass gain.

ovis··on NASA Study: Mass Gains of Antarctic Ice Sheet Greater Than Losses
I agree with much of what you write, particularly the first three paragraphs. I wanted to make a few comments.

Yes, climate science deals with problems are that more difficult to solve with the "hypothesis->experiment->new hypothesis" approach that works so well for some other fields. People try their best with models and paleo- studies, but it's far, far from ideal. As a result, the IPCC must use other methods to assess consensus.

Many of the fundamental tenets of climate science are reproducible and falsifiable (absorption spectra of the atmosphere with different pollutants; the effect of acidification on marine carbonates; the ways in which glacier ice deforms and flows, as examples). But figuring out how they fit together on the scale of a planet is an enormous challenge.

Ultimately, there's a lot of uncertainty, which some have argued is intrinsic to the problem. What if we can't answer your train of questions in a reasonable amount of time with yes/no? If the answers are variations on "maybe" (they usually are), how should we manage risk? Does ethics come into play?

Finally, there is a pretty rich history of people (scientists included) inadvertently radically-changing their environments because the time scales are too long to grasp.

ovis··on NASA Study: Mass Gains of Antarctic Ice Sheet Greater Than Losses
I'm not sure how it could be considered a "mark against [...] AGW". One thing to keep in mind is that temperatures in East Antarctica are nearly always below freezing, so rising temperatures don't have much of an effect at first on mass loss[0].

From the article, the ice accumulation seems to be from increased snowfall during the Holocene. How this will ultimately be affected by warming isn't clear, but for now it seems somewhat orthogonal to AGW.

[0] Instead, most mass loss from Antarctica appears to be at the margins and driven by ocean processes.

ovis··on NASA Study: Mass Gains of Antarctic Ice Sheet Greater Than Losses
There are a few important processes that cause observed sea level rise. In order, with IPCC estimates on the magnitude:

- The ocean expands as it warms, which is the single largest contributor right now (1.1 mm/yr).

- Mountain glaciers are next, as they are very rapidly melting (0.76 mm/yr, with large uncertainties).

- Then Greenland, as mentioned by another poster (0.43 mm/yr).

ovis··on Why NASA Needs a Programmer Fluent in Fortran and Assembly Language
I've found Metcalfe, Reid, and Cohen[1] to be a fairly clear refresher on Fortran. Unfortunately, the build tools I use don't support many of the nicer features like user-defined types.

[1] https://books.google.ca/books/about/Modern_Fortran_Explained...

ovis··on Show HN: CADWOLF – Web-based math and engineering platform
I think this is very interesting, although I'm still trying to understand it. So excuse my ignorance, but how does this differ from open source projects such as Jupyter, Sage, and JuliaBox?
ovis··on Visualizing the Shrinking Sea Ice
One contributor is an increase in the amount of heat flowing from the North Atlantic into the Arctic ocean. Another factor is that the volume of old, thick ice has been declining as well, so the ice present in 2015 is younger than what was present in 2000. This ice is less likely to persist through the summer melt season.
ovis··on Visualizing the Shrinking Sea Ice
Also, the sea ice itself is much fresher than the ocean, so it's melting temperature is much closer to 0C than you might expect. And even the distinction between freezing seawater and freshwater at the surface is less then 2C, so it's not a huge difference.

I can't think of an obvious way that acidification factors in, unless there are important connections between sea ice and sensitive biological factors (I'm not aware of any, but I'll wait to be corrected).

← PreviousPage 2 of 4Next →