Agreed. The bookshelves have been full of terrible sci-fi for over a decade now. I have given up on the book sellers and the awards (Hugo, Nebula etc). I do read sci-fi almost every day though still. Almost all of it from Kindle Unlimited.
This article could probably be summarized as "People don't read anymore".
There is a gap between the title of the article and the contents. Starts out with weather forecasting is improved, but spends most of the article talking about how poor people and countries have other things to spend their money on than forecasting weather.
No, but I breathe oxygen radicals 24 hours a day and so do you. A study about the free radical content of the air in a bag of Lay's Potato Chips is pretty pointless despite it being proven to cause cell death.
I appreciate the response, this is good to know. I apologize for not being clearer but what I'm specifically looking for is the "GeoColor" full-disk imagery [0], which I don't think is archived anywhere (beyond a couple weeks worth). That imagery is generated from the raw data you linked, but sadly the code isn't publicly available.
There is a good amount of material around the techniques used to generate the GeoColor images [1][2] but understanding it well enough to implement it is a little above me.
If any HN'er can code up something like this, I'd definitely be interested in talking with them.
Do you have any kind of "geospatial spreadsheet" processing software to take in raw sat datasets and an algorithm for combining bands and applying filters to create an output?
Eg: the commercial ERDAS Imagine or other products?
I imagine it would - it seems to be a modern version evolved from ERMapper of the late 1980s | early 1990s that did exactly this kind of work.
ERMapper was a program that could pipe in many raw input multi channel data files from satellite | aircraft | other sources along with an "algorithm" file that described what to do.
One part of the task is to geo mosaic and merge data from seperate files together so that per output location there is an awareness of what cells are local to that output, the cell dimensions, type of data, channel, etc.
The algorithm part is to (say) combine data from channels 3, 7, and 9 using a particular filter and to adjust (say) for rayleigh scattering.
Such programs are geospatially aware spreadsheet programs that pipe in raw data an pipe out "cooked" or processed data.
When the GeoColor scientific papers you linked talk about the filtering operations they do on channel data they very likely are not "coding" in the sense of writing C code - but constructing "algorithms" by specifying which mathematics operations to perform on input channel data and writing that out in the algorithm file*.
* Which a pre existing program (perhaps ERDAS Imagine) reads and applies to it's input data, the raw captures from the satellites.
Yes, I wrote such things 35+ years ago, no I'm not in the field today and don't have a copy of ERDAS Imagine.
I'm not sure what the state of art for open source geomosaic software is these days - something like
The real estate and banking industry with a host of grifters caused 2007-2008 crash....then we elected a direct representative of this group as president 10 years later. The problem is deeper than poor economic policy.
This article could probably be summarized as "People don't read anymore".